Lenski Peasant - and - Slave - in - Late - Antique - North

Late Antiquity in
Contemporary Debate
Edited by
Rita Lizzi Testa

Late Antiquity in Contemporary Debate
Edited by Rita Lizzi Testa
This book first published 2017
Cambridge Scholars Publishing
Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Copyright © 2017 by Rita Lizzi Testa and contributors
All rights for this book reserved. No part of this book may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without
the prior permission of the copyright owner.
ISBN (10): 1-4438-4308-3

ISBN (13): 978-1-4438-4308-9
International Committee of Historical Sciences –

Comité International des Sciences Historiques
TABLE OF CONTENTS
Introduction ............................................................................................... vii

Rita Lizzi Testa
Historiography 1
Chapter One ................................................................................................. 2

Empire and Aftermath
Clifford Ando
Chapter Two .............................................................................................. 15

Crisis, Transition, Transformation: The End of the Roman World
and the Usefulness of Useless Categories
Pablo C. Diaz
Methodology: Sources and Periodization
Chapter Three ............................................................................................ 38

Transformation and Transition in the Art of Late Antiquity
Jutta Dresken-Weiland
Chapter Four .............................................................................................. 56

Defining Late Antiquity through Epigraphy?
Ignazio Tantillo
Case Studies
Chapter Five .............................................................................................. 80

Reddite quae sunt Caesaris, Caesari: The Late Roman Empire
and the Dream of Fair Taxation
Gilles Bransbourg
Chapter Six .............................................................................................. 113

Peasant and Slave in Late Antique North Africa, c. 100-600 CE
Noel Lenski
vi Table of Contents
Chapter Seven.......................................................................................... 156

What is Geo–Ecclesiology: Defining Elements Applied to Late Antiquity
(Fourth–Sixth Centuries)
Philippe Blaudeau
Historiography 2
Chapter Eight ........................................................................................... 174

The Historical Path of “Late Antiquity”: From Transformation
to Rupture
Jean-Michel Carrié
Concluding Remarks: The Birth of a New Short Late Antiquity .............. 215
Hervé Inglebert
Contributors ............................................................................................. 228

CHAPTER SIX
PEASANT AND SLAVE IN LATE ANTIQUE

NORTH AFRICA, C. 100-600 CE
NOEL LENSKI
One of the most enduring questions in the study of Late Antiquity is that
of labor organization and land tenure. From Marx, to Fustel de Coulanges,
to Weber in the nineteenth century, through Rostovtzeff, Mazzarino,
Štaerman, and Finley in the mid-twentieth and up to the present with Lo
Cascio, Vera, Carriè, Banaji, Wickham, Sarris, Grey, and Harper, the
question has preoccupied some of the greatest historical minds.1 Did Late
Antiquity open a new chapter in the organization of labor relations? Or
was it simply more of the same? Did it remain a period whose economic
organization was characterized by what Marx termed the “slave mode of
production” and Finley repackaged in sociological terms under the concept
of a “slave society”? Or did it, as both Finley and Marx agree, give way to
some form of bound dependency short of enslavement, a condition the
mid-twentieth century was content to label “serfdom”? Based primarily on
evidence from Egypt, Banaji and Sarris have recently restated the case for
the rise of a new mode of tenant labor organization with the rise of great
estates populated by tenant laborers working on long-term (essentially
permanent) rental contracts known in the sources generically as coloni.2
Even more recently, however, Harper has taken a very different position
and made the case that late Roman society remained a Finleyan “slave
1
Weber 1891; Fustel de Coulanges 1894; Rostovtzeff 1910; 1957; Mazzarino
1951; Štaerman 1957; Finley 1973; 1980; Lo Cascio 1997; Vera 1983; 1987; 1992-
1993; Carrié 1982; 1983; 1997; Banaji 1997; 2001; Wickham 2005; Sarris 2006;
Grey 2011; Harper 2011. I should like to thank Monica Hellstrom for her helpful
discussion of this paper as well as Dennis Kehoe and Domenico Vera for their
careful readings and useful advice.
2
Banaji 2001; Sarris 2006.
114 Chapter Six
society” throughout the long fourth century.3 Engaging in particular with

Marxian models that emphasize a transition from slavery to bound
tenancy,4 Harper has asserted that agricultural surplus continued to be
produced for the Roman elite primarily through the exploitation of slave
labor into the fifth century. In what could be termed a “capitalist model,”
built on Finleyan assumptions about the importance of markets for the rise
and flourish of a “slave society,” Harper contends that only with the
political collapse of the Roman empire in the fifth century did slavery
falter on the shoals of economic stagnation. As the market for the surplus
production dried up, the demand for large-scale slave operations shrank
and with it went Rome’s slave society.
In order to make his case, Harper must sidestep the abundant evidence
for agricultural tenancy throughout the imperial period and for its gradual
transformation into a more slavelike condition with the binding of coloni
to the estates on which they were born. Rather than discuss this evidence
head on, Harper falls back on the work of Jean-Michel Carrié, who has
argued strongly against the existence of the “colonate” as any coherent
juridical status, contending instead that it represents a modern “mythe
historiographique.”5 Although Carrié’s arguments have been reinforced in
recent work by Grey, the thesis remains controversial. Thus Harper’s
contention that reference to Carrie’s work suffices to replace serious
engagement with the detailed and abundant source pool on tenancy leaves
his argument open to question.6
In this brief study I return to the question with a focus on just one
geographical area, the North African Maghreb, a region of central
importance to Harper’s thesis because of the rich storehouse of circumstantial
material on agricultural production available in the corpus of Augustine.
My goal is to reopen the question of agricultural labor in this territory and
particularly to weigh the evidence for the relative importance of free,
bound, and slave labor in generating surplus for the regional and imperial
3
Harper 2011, 144-200, 497-509.
4
Especially the work of Vera and Giardina; cf. Giliberti 1999; Rosafio 2002.
5
Carrié 1982; 1983; 1997.
6
Harper 2011, 153–55: “Suffice it to say that Carrié’s initial critique forever
undermined the idea that the colonate was a ‘replacement’ of the slave system or
that it created an intermediate serf–like status between slavery and freedom.” See
also Harper 2012, 169: “Second, one of the central components of the narrative of
transition from ancient slavery to medieval serfdom has disappeared, at least in
anything like its classic form: the colonate... Indeed, just as the colonate is a
‘historiographical myth,’ so too the conquest thesis is an ‘economic myth’.”
Peasant and Slave in Late Antique North Africa, c. 100-600 CE 115
elite. I explore the question over the longue durée, for only thus can we
hope to determine whether there was in fact a transformation in labor
relations during Late Antiquity, or any other period for that matter. To
anticipate my conclusions, I will argue based on what should represent a
fairly comprehensive examination of available sources that, even if there
may have been a “slave society” elsewhere in the Roman Empire in the
fourth century, there is no solid evidence that this was the case for North
Africa at any period between the first and sixth centuries CE. Indeed,
while plenty of evidence for agricultural labor in the region exists, most of
it points to the use of tenants for the generation of surplus on the estates of
the emperor and the political and economic elite in all periods of antiquity
with the possible exception of the second and fifth centuries—the later
only in the post-Roman period of Vandal occupation. Second, I hope to
demonstrate that, although tenancy remained a prevailing—usually the
prevailing—mode of rural labor organization, it did in fact shift in Late
Antiquity with the rise of the bound colonate. The net effect was a marked
reduction in the freedom of farm laborers, who had formerly been given
considerable sway in the management of their tenancies but lost this to a
welter of state regulation. This alteration to traditional forms, well attested
in the writings of Augustine, then saw further shifts with the arrival of the
Vandals, who appear to have relied much more heavily on slaves. With the
East Roman reconquest of North Africa, the imperial fiscal administration
made a half-hearted effort to reimpose a bound colonate but essentially
faltered in the grip of the centripetal forces that always made the binding
of tenants difficult, and particularly in a region where longstanding local
custom had favored freer forms of tenancy since the early empire. In this
sense, Late Antiquity did indeed witness a period of transformation in the
organization of agricultural labor, but not one that was unidirectional. The
heavy, if hardly predominant, reliance on slave agricultural labor in the
second century gave way to the much more intensive use of bound tenant
labor under the dirigiste state of the fourth century, and in turn to the
opportunistic exploitation of slavery in the fifth, only to revert to patterns
of bound tenancy, albeit with limited success, in the sixth. Insofar as this is
true, a Finleyan “slave society” is not to be found in any period of Roman
rule in North Africa, least of all in the long fourth century.7
7
Vera 1988 already reaches similar conclusions to those presented here. So too
Lassère 2015, 205-6, 261-280; Whittaker 1978; 1980; Garnsey 1998, 473: “North
Africa was a slaveowning rather than a slave society.”
116 Chapter Six
Natural parameters––land and water, large estates

and small-scale competition
There were certain fixed factors that governed the organization of
agriculture and land tenure in most of North Africa throughout the Roman
period. Chief among these was geography and its relationship to climate
and water supply. Roman North Africa essentially corresponded to the
territory known with the Arabic designation Maghreb (“West”), stretching
along the south-western coast of the Mediterranean between Tingis
(Tangiers) in Mauretania and Lepcis Magna (Lebda) in Libya. This was a
landscape dominated by the long ridges of the Atlas and Aurès Mountains,
paralleling the coast from south–west to north–east between modern
Morocco and Tunisia. These ridges, whose heights were located between
200 and 400 km inland, simultaneously captured precipitation and
fragmented the geography into a series of fertile valleys (suitable for
cereal agriculture), surrounded by hillsides (whose lower slopes supported
fruit and olive production). Along the central north coast, in what is today
Tunisia and eastern Algeria, for a distance of between 200 and 300 km
inland, average rainfall exceeded 400 mm annually, making cultivation
possible without irrigation. Here ceraliculture flourished already in pre–
Roman times. It was given further impetus by Roman centuriation in the
fertile valleys of the seaward flowing watersheds, which resulted in the
creation of orderly plots that were parceled out to Roman colonists,
primarily in the first centuries BCE and CE. For another 100–200 km
southward an isohyet of 200 mm of annual rainfall allows for irrigated
agriculture, which gradually took off under the stable security situation
provided by the Romans, reaching an apex in the third century CE. This
more marginal steppic land, as for example in the territory of western
Byzacena, tended to favor oleiculture rather than grain production,
although olives as well as fruits (figs, dates, and wine grapes) were
produced all across the Maghreb, particularly at the edges of arable tracts.
Beyond this to the south was desert, accessible only to pastoralist nomads,
who were themselves very much a part of the region’s economy as they
entered and exited arable territory, interacting with the agriculturalist
neighbors as occasional laborers, traders, and raiders.8
Agriculturalists have tended to cluster their habitations around rivers,
wadis, and springs and have always worked to harness and husband
8
More on the landscape of North Africa and its effects on habitation and
agriculture at Lassère 2015, 21–26, 201–43.
precious supplies of water using dams, reservoirs, and conduits. This

hydrological situation has favored those who were able to concentrate
capital, even as it also fostered competition for control of limited water
and limited parcels of land with access to it. Closer to the desert steppe, it
has increased interaction with nomads, further conditioning the
agglomeration of populations into compact settlements for purposes of
defense.9 Already in pre-Roman times this nexus of geographic and
climatological factors gave impetus to the growth of large estates, which
could concentrate and control water resources and provide security against
nomadic indigenes.10 Such latifundia retained their importance in the
period of Roman imperial rule, when large landholdings are alluded to in a
famous passage of Pliny the Elder claiming that, “latifundia have
destroyed Italy, but also the provinces; six estate-holders owned half of
Africa when Nero murdered them.”11 A similar situation is described in
the early fifth-century surveyor Agennius Urbicus:
But they [i.e., disputes between private landholders and civic communities]
often occur in the provinces, especially in Africa, where private individuals
have estates no less extensive than the territory belonging to a civic
community (res publicae territoria); indeed, many estates are far bigger
than territories. Moreover, private individuals have on their estates a not
insubstantial population from the lower orders (populum plebeium), and
villages (vici) scattered around their country house rather like towns
(municipia).12
9
Whittaker 1978. On tensions between nomads and sedentarists over territorial
control, see CIL 8: 8369 = ILS 5961with Kolendo 1991, 32. See Mattingly et al.
2013 for a survey of fortified farms and villages in Africa.
10
The Numidian vicus Phosphorianus at Aïn Melouk near Thibilis controlled
some 4425 ha, cf. Desanges 1989. Shaw 1992, 92 argues that at Lamasba (̓Ain
Merwâna), also in Numidia, a dozen families controlled 72% of the land recorded
at CIL 8: 4440 = 18587 = ILS 5793 (220s). For archaeological evidence of large–
scale property concentration in the Segermes Valley, see Dietz et al. 1995,
11
Plin. Nat. Hist. 18.6.35: latifundia perdidere Italiam, iam vero et provincias; sex
domini semissem Africae possidebant, cum interfecit eos Nero; cf. Sen. Ep.
19.5[114].26.
12
Agennius Urbicus, De controversiis agrorum: Inter res p. et privatos non facile
tales in Italia controversiae moventur, sed frequenter in provinciis, praecipue in
Africa, ubi saltus non minores habent privati quam res p. territoria; quin immo
multi saltus longe maiores sunt territoriis: habent autem in saltibus privati[s] non
exiguum populum plebeium et vicos circa villam in modum municipiorum
(Campbell 2000, 42–43 with 349–50 n. 56). On the date of Agennius Urbicus, see
Campbell 2000, xxxi–xxxiii.
118 Chapter Six
In Africa, then, private estates were often as extensive as neighboring

towns and even stood in competition with these as centers of territorial and
political autonomy.13 Moreover, as the Pliny passage implies, Africa was
notorious for its especially high concentration of imperially owned
properties. Using the quite precise figures mentioned for cultivable estates
in Africa Proconsularis and Byzacena in a law of 422 CE, Claude Lepelley
has estimated with some precision that the imperial res privata controlled
approximately 1/6 of the cultivable land of both provinces in this period,
and the same is likely to have been true in earlier centuries as well.14
It is important to note that the archaeological and epigraphic evidence
adds considerable subtlety and complexity to the monolithic impression of
outsized estates left by the texts.15 Particularly in those areas that could
support wet farming, we have evidence for extensive centuriation into
smaller plots and epigraphic as well as archaeological testimonia to a vast
multiplicity of individually named farms, both private and imperial.16
Even in more marginal landscapes, archaeological survey has indicated the
regular recurrence of tessellation into relatively compact estates which
competed with one another for access to resources and to maximize
surplus output through the investment of capital and labor into
improvements like olive presses and grain storage silos. It remains a
subject of ongoing debate whether these opposing indications (of large and
small scale operations) in the source pools point to a shift over time (as
Africa went from a labor deficit in the early empire to a labor surplus by
the early third century)17 or perhaps simply two versions of the same story
(large–scale landholders co–existing alongside smaller free–holders and
maintaining control over holdings scattered across a dispersed landscape).18
13
See also Vita Melaniae Latina 21 (Laurence pp. 194-95): quae possessio maior
etiam erat civitatis ipsius (speaking of the estate of the Valerii near Thagaste).
14
CTh 11.28.13 with Lepelley 1967. See also Crawford 1976, 57-59 for a
provisional list of known imperial estates.
15
Fentress et al. 2004, passim, esp. 160: “In spite of a tendency to agglomerate, the
African countryside remained marked by a highly differentiated settlement pattern,
with villages, small farms and towns surviving beside the villas of the urban elite.”
On a micro-regional level, Leveau 1984, 281–397 explores the variety of
productive strategies and their correspondence with landscape and labor in the area
around Caesarea. Invaluable is the survey of North African archaeology at
Mattingly and Hitchner 1995.
16
For a sampling, see Lengrand 1996, 112–16.
17
So Lassère 1977, 647–62.
18
Picard 1959, 64, 373; cf. Lassère 2015, 20–22.
These competing narratives are in many ways emblematic of a broader

story of competition between labor and management that played itself out
continuously from the first through sixth centuries CE. Small freeholders
and tenant farmers struggled to assert control over the land they and their
families had been farming for generations and the surplus product it
produced, even as larger landowners (private and imperial) struggled to
maximize control over the land and water resources to which they laid
claim, resources developed and farmed for them by slaves, tenants, free
wage laborers, and even small freeholders leasing land from larger estates.
In other words, the simultaneous impression of land and resource
concentration and fragmentation may be an artefact of an ongoing tug–of–
war between freeholders, landlords, and tenants, each with competing
claims to resources and output. The net result was what Domenico Vera
has referred to as a jigsaw-puzzle of arrangements on imperial as well as
private holdings consisting of a variety of types of land under a variety of
types of cultivation by a variety of laborers and for a variety of owners.19
When it comes to describing the labor pool in the region, there is
evidence for a similar degree of variability. Nevertheless, where good
documentation for the labor regime on rural estates is available, it
indicates that this was overwhelmingly free, whether this meant free
landowners on small privately held farms, free tenants renting parcels
from mid-sized landholders, or free coloni operating on larger estates. This
is certainly true of the evidence for imperial estates, which are reasonably
well documented epigraphically for the high empire, and the sources we
have for labor on private estates from the later Empire also point to a
general tendency toward the use of free, albeit legally bound, coloni. This
is not to deny that other forms of land tenure and labor organization
existed across this variegated landscape. In the steppe zones on the desert
edge, animal husbandry and nomadic pastoralism were common, both
among non-Romanized Berbers and among subjects of the Roman Empire.
Here, however, the personal status of such herders is difficult to determine,
and there is no evidence that herding generated a significant surplus for the
landholding elite. We also have evidence for free seasonal laborers, most
famously the “Mower of Mactar,” whose self-congratulatory epitaph
stands as a pointed reminder that free wage and contract labor constituted
an important component of the overall economy in a region where
19
Vera 1988, 986. Similarly for other parts of the empire, cf. Hickey 2007; 2012,
62–89.
120 Chapter Six
seasonality played a central role in the agricultural labor cycle.20 Finally,

there is, as we shall see, irrefutable evidence for the use of slave labor in
the first six centuries CE. Although it is never as high-resolution and
abundant as the evidence for tenancy—in large part because of the
widespread tendency of all source types to underrepresent slaves—it
should serve as a reminder that labor relations were as variegated as
landholding in the region. What we do not have, however, is any clear
indication that slave laborers were central to the production of agricultural
surplus for the elite. Instead, the overwhelming majority of the evidence
indicates that long–term dependent or semi–dependent tenants were the
primary cultivators of the medium and large-scale estates from which
landholding aristocrats extracted their wealth in North Africa.
First and second centuries

Already in the Carthaginian period, agricultural labor was organized
around fortlets (castella) using large–scale tenancy arrangements. These
were simply carried forward with the rise of Roman hegemony. By the
first century CE both the emperor and Roman senatorial aristocrats
controlled huge amounts of North African land which, perforce, they
managed as absentees through conductores––management level lessees
who took five–year contracts on extensive farmsteads that consisted of
numerous holdings parceled out to individual tenant–farmers (coloni)
expected to pay rents in kind directly to the conductor, who generally also
controlled a pars dominica that he exploited directly. Local landholders
appear to have followed a similar model, exploiting their estates through
tenants or sometimes slaves whom they managed either through conductores
or more directly through procuratores, actores or vilici (managers, some
of them slaves, to whom they entrusted estates for supervision).21
Absenteeism was the norm, requiring the availability not just of an
adequate labor pool but also of reliable supervisors operating with equally
reliable tenure arrangements to maximize efficiencey and, above all,
minimize risk.
In the second century CE, we have excellent evidence for such
arrangements in a series of six inscriptions discovered in the Bagrada
20
CIL 8: 11824 = CLE 1238; cf. Shaw 2013, 48–92, with bibliography and further
references. Banaji 2001, 190–212 makes a case for the centrality of wage labor in
early Byzantine Egypt, but see now the critique of Freu 2013.
21
Whittaker 1978, 335–41.
(Medjerda) River Valley in the province of Africa Proconsularis.22 All are

related to the management of imperial estates, although there is reason to
believe that the arrangements they describe were characteristic of the
region more broadly. The earliest and most detailed, dating from late in the
reign of Trajan (c. 117 CE), relates to a landholding called the Villa
Magna Variana at the modern site of Henchir Mettich (HM).23 Its text
makes it clear that contractual relations between conductores and coloni
on this estate were regulated by a normative code termed the Lex
Manciana. Ever since the work of D. Flach and D. Kehoe in the 1970s–
1980s, scholars have generally accepted that this was originally a private
law arrangement which came to be applied to estates of the fisc that the
emperor had acquired and continued to manage under its provisions.24 The
HM inscription represents a response to a petition lodged by the coloni of
the Villa Magna asking for clarification on rental rates or shares (partes)
and the management of fallow marginal land (subseciva) that had been
brought under cultivation by the tenants. By the terms of the inscription,
rental rates were to be regulated at one–third shares for most crops, one–
quarter for some others, to be paid in kind to the conductor. In addition,
tenants were required annually to perform six days-worth of labor service
(operae) on the pars dominica under the direct control of the conductor.
Some have speculated that conductores generally cultivated these direct
concerns using slave laborers, although there are no clear indications of
this in the epigraphic record, and the six days of corvée labor required of
the various coloni probably would have minimized the need for much
additional labor input, depending of course on their size.25 Less
speculatively, and more important for the overall productivity of the estate,
the HM inscription provides that coloni who succeeded in bringing
marginal land under cultivation would gain an ownership claim in the
22
See Flach 1978 and Kehoe 1984a; 1985; 1988, 28–70 for a complete set of
editions and translations; cf. Flach 1982 and Kehoe 1988 passim for analysis.
Kolendo 1991[1976], 34-45 remains useful; and see Hobson 2015, 54-61.
23
I follow the text established at Kehoe 1984a, 198–201; 1988, 29–38. See also
Flach 1978, 476–84. Kehoe’s reading now supersedes CIL 8:25902 = Riccobono
FIRA2 no. 100.
24
Kehoe 1984a, 202–3; 1984b; 1988, 48–55; Flach 1982, 447–46; De Ligt 1998–
1999, 219–20.
25
For the likelihood of slave laborers on the domanial estates, see already Schulten
1896, 88-91; cf. Vera 1992, 473; Wickham 2005, 273-74. Slaves are mentioned in
the final lines of column IV of the Henchir Mettich inscription as guardians, but
the inscription breaks off before further details can be gleaned.
122 Chapter Six
lease rights (usus proprius) to that newly productive soil which gave them
an exclusive claim to cultivate it and to pass these rights on to their
successors through inheritance. It even gave coloni a grace period of five
years (for vines) and ten (for olives) to bring such new plantations to full
productivity before they owed rent on them.26
This second proviso obviously encouraged the development of
marginal land and thus increased the overall productivity of an estate. It
also encouraged entrepreneurialism among coloni, who gained quasi-
ownership rights from their labors and were thus incentivized to collude
with the imperial fisc––the ultimate owner of the land––in increasing
output. With the HM inscription and indeed all of the Bagrada Valley
texts, the emperor seemed to be further supporting this relationship by
circumscribing the rights of his middleman lessees (conductores) to usurp
the profits of the coloni or otherwise abuse them. By preventing the more
powerful agent in this binary rental arrangement from gaining the upper
hand, the emperor was fostering primary producers and encouraging the
full exploitation of estates that were far too distant and numerous for him
to manage personally.27 This sort of savvy management strategy is only
thinkable in a free labor environment, a fact that must have incentivized
reliance on tenants rather than slaves, who could only be acquired and
maintained with significant capital outlays and whose labor productivity
could only have been increased with coercive strategies that degraded their
physical capacity and compromised their reproductivity.28
A major question that remains with regard to the Bagrada Valley
inscriptions is whether both of the main spheres of regulation––that
concerning rental rates and that encouraging the cultivation of marginal
26
Schubert 2008 offers a radical new reading of the Henchir Mettich inscription
that assumes the provisions on marginal land deal with the sedentarization of
nomads and that the related inscriptions (discussed below) then show an evolution
as these sedentarized tenants faced increasing abuse in subsequent generations.
The thesis is worthy of consideration, but I remain more convinced by Flach’s and
Kehoe’s interpretations.
27
This is outlined brilliantly at Kehoe 1988, esp. at 71–153.
28
Pace Harper 2011, 159–60: “Tenancy inherently discouraged capital investment,
partly because tenants had little reason to invest in long–term improvements and
because the principal cash crops of the Roman empire, grapes and olives, take so
long to mature.” In a general way, Harper’s model is based on the unproven
assumption that Roman landholders’ primary goal was to maximize return on
investment, but much better attested were strategies that minimized risk while
guaranteeing steady returns, see Kehoe 1997; cf. Vera 1983.
land (subseciva)––were part of the original lex Manciana, or whether the

lex Manciana only regulated the first and the provision for the cultivation
of subseciva in the HM inscription was a new addition to this protocol first
introduced by the very text recorded there. Luuk De Ligt has argued
convincingly that both provisos were part of the original law, which
simultaneously set (relatively low) rental rates on land cultivated by coloni
and encouraged the entrepreneurial exploitation of marginal land and its
long-term maintenance as quasi-property by tenant–farmers.29
The process of bringing unused land into production with cereals, fruit
trees, and olives was then further encouraged by a “law of Hadrian on
uncultivated lands” (lex Hadriana de rudibus agris) which broadened the
schedule of lands eligible for new cultivation to include centuriated
territory which had been abandoned. We have long known of its
promulgation in Africa through two further inscriptions from the Bagrada
Valley at Aïn-Djemala and Aïn-Wassel.30 The first was issued under
Hadrian, but the second dates to the reign of Septimius Severus, indicating
that the order was generalized across time. A third epigraphic attestation,
published in 2000 from Lella Drebblia near ancient Thugga, confirms that
the lex Hadriana de rudibus agris was truly widespread in its application
across the region.31 Indeed, it has been argued that the lex Hadriana was
not limited to Africa but was implemented more broadly across the
empire.32 The African inscriptions attesting to it seem then to have grafted
its provisions into the pre-existing matrix of Mancian tenure arrangements
customary in the region in order to create a system that was extremely
generous in distributing possessory rights to ambitious planters and thus
encouraging cultivation. The combination of lex Manciana and lex
Hadriana de rudibus agris thus pioneered and propagated an ingenious
way simultaneously to dampen the power of conductores and increase the
productivity of coloni. Not only did they set reasonable rental rates, but
they also granted perpetual proprietary claims for new plantations of vines,
fruits, and olives that encouraged the expansion of market agriculture by
small-scale tenants.33 Moreover, as written protocols, both could be and
29
De Ligt 1998–1999.
30
Kehoe 1984b, 159–62; 1988, 55–59; cf. Flach 1978, 484–89. Kehoe’s reading
supersedes CIL 8: 25943; 26416 = Riccobono FIRA2 no. 101–102.
31
AE 2001, 2083 = De Vos 2000, 35.
32
Lassère 1977, 298-99; cf. Scholl and Schubert 2004.
33
For the variety of fruits cultivated in the Maghreb under Roman rule, see Lassère
2015, 212–16, with earlier bibliography.
124 Chapter Six
apparently were cross–applied throughout North Africa, where they

apparently governed both imperial and private holdings.34
The enduring impact of this framework is confirmed by the fact that
the general terms of the lex Manciana and the lex Hadriana de rudibus
agris formed the regulatory basis for tenancy arrangements mentioned in a
series of petitions to the emperor Commodus in the 180s. The most
extensive of these comes from the Saltus Burunitanus (Suk el-Khmis), 50
km west of the Villa Magna Variana. It records a petition from the coloni
of this estate complaining that their rent shares and corvée labor
obligations had been arbitrarily increased by a procurator who had gone
so far as to unleash soldiers to arrest and even beat some of the tenants––
some of them Roman citizens. In his rescript, Commodus upheld the
claims of the coloni and forbade further abuse, to what effect we can no
longer say.35 Two further petitions to Commodus, both extremely
fragmentary, point to similar tensions between managers and coloni. One
at Aïn Zaga preserves only the heading of the petition, but the second,
from Gasr Mezuar, preserves enough syntax to identify further complaints
about radically increased corvée obligations as well as increases in rent
shares on fruit trees.36
This collection of late third century material indicates a growing
tension between colonus and manager, a situation that would continue to
escalate to the disadvantage of the colonus. Still in the 180s, however, the
emperor strove to uphold the inherited claims of his tenants to limitations
on their rent and labor obligations and to impose an interdict on the
arbitrary abuse of their persons. As we shall see in the next section, this
tendency of the emperor to defend primary producers in the face of
procuratores and conductores appears to have endured into the early
fourth century, after which the emperor began using the force of law
increasingly to the advantage of managers (imperial and private) over
against the interests of free tenants. Even so, the lex Manciana played a
role in guaranteeing rights to land tenure as late as the fifth–century, when
the Albertini Tablets, found some 200 km south of the Bagrada River (and
thus in a very different agricultural context and, moreover, in a private
34
Schubert 2008, 253-54.
35
Kehoe 1988, 64-69; cf. Flach 1978, 489-92. Kehoe supersedes CIL 10570;
14464 = Riccobono FIRA2 no. 103.
36
CIL 8: 14428 = ILTun 1220 (Gasr Mezuar). CIL 8: 14451 (Aïn Zaga).
tenancy arrangement) continue to record the claims of tenants to the

cultivation of their fields under the lex Manciana.37
None of this is to deny that there is good evidence of the presence of
slaves in the region and their use in agriculture.38 Slave-like figures appear
regularly on mosaics of the second through fourth centuries in both
domestic and agricultural roles, even if it is extremely difficult to
determine in any given instance whether a particular image represented a
slave rather than a free dependent laborer.39 Indications of slaves in
agriculture are also present in the Apology of Apuleius. When describing
the Tripolitanian estates of his wife Aemilia Pudentilla, he reports that she
controlled “extremely fertile lands and a great and richly decorated house
as well as a huge amount of wheat, barley, wine, olives, and other fruits,
and also scarcely less than 400 slaves, and even more herd animals.”40
Although the passage does not confirm that these slaves worked the land,
this conclusion seems inevitable given the numbers at Pudentilla’s
disposal. Indeed, earlier in the same text Apuleius fends off his brother–
in–law’s ridicule at the low number of slaves he himself possessed by
pointing out that, while some Africans deployed slaves on their farms,
others used free laborers41—a confirmation of the overall picture presented
here that slaves were a choice, not the choice for agricultural labor on the
part of elite proprietors.
In fact, in many instances where slaves can be found in rural North
African settings, they can be set alongside free tenants, who tend to
outnumber them quantitatively and qualitatively as the primary producers
of agricultural surplus. Thus Stéphane Gsell assembled some 30
epigraphic attestations for the use of slaves in managerial positions on
37
See below n. 106. Hobson 2015, 57-61, 146, 157-60 is less convinced of the
general applicability of the Lex Manciana outside the Bagrada Valley but has
difficulty explaining its regular appearance in the Albertini Tablets.
38
Leveau 1984, 20-22, 82-83, 98-100, 149-53 has shown how the epigraphic
record of Caesarea (Cherchel) indicates a relatively high percentage of slaves in
this urban environment in the first century BCE through third century CE, although
he cautions against assuming this picture applied to the countryside.
39
Blazquez 1998.
40
Apul. Apol. 93.3–5: fructuosissimos agros et grandem domum opulente ornatam
magnamque uim tritici et ordei et uini et oliui ceterorumque fructuum, seruos
quoque haud minus CCCC, pecora amplius neque pauca neque abiecti pretii
donaret.
41
Apul. Apol. 17: Ego adeo servosne tu habeas ad agrum colendum an ipse
mutuarias operas cum vicinis tuis cambies, neque scio neque laboro.
126 Chapter Six
rural estates, but these are far outnumbered by the 66 inscriptions

assembled by Klaus-Peter Johne, Jens Köhn and Volker Weber that attest
to the presence of coloni on North African farmsteads.42 The disparity in
numbers is unlikely to be representative of the relative differences of these
two groups in the North African countryside, but it does foreclose any
argument that North Africa of the high empire was a “slave society,” i.e. a
society in which members of the elite looked to slave labor as the primary
or even exclusive manpower pool for generating income. Similarly, one of
our best testimonia to the use of slaves in North African agriculture comes
in a passage from the second–century jurist Scaevola where he describes a
woman who left a legacy of her African farm “outfitted with all its effects
and with its slaves and the rents outstanding from its coloni.”43 The
passage is striking for its implication that the estate was as reliant on free
tenants as it was on slaves for its workforce. Indeed, one wonders if we are
not dealing here with a private law instance of precisely the sort of land
tenure structures observed in epigraphic attestations of the Mancian law,
with slaves tending a domanial home-farm while coloni cultivate the
surrounding tenancies. Whether or not this is the case, Scaevola’s
testimony reinforces the impression that slaves and tenants operated side
by side in African agriculture.
Slaves were thus by no means the only, nor even the primary,
producers of agricultural surplus in the high empire. In fact, when we
examine the most circumstantially detailed sources from the period––the
Bagrada Valley inscriptions––we cannot escape the impression that
tenants predominated as the generators of agricultural wealth for large-
scale landholders. While these inscriptions are directly relevant only to
imperial estates, there are strong indications that they reflect a pattern of
estate management characteristic of private estates as well. Large
landholders benefitted from the flexibility of part slave / part free work
forces, each with its own drawbacks but––given the ingenious equilibrium
struck by the Mancian tenure custom––each likely to have been effective
at guaranteeing steady profits in a region characterized by topographical
and climatological variability.
42
Gsell 1932, 402-8; Johne, Köhn and Weber 1983, 372-410. Carlsen 1991 also
concludes that the epigraphic attestations of vilici and actores indicate a heavier
reliance on tenants than slaves in the High Empire.
43
D 33.7.27.1: “Fundum Cornelianum Titio ita ut est instructus cum omnibus
rebus et mancipiis et reliquis colonorum dari volo.” For a similar mix of tenants
and slaves see Aug. Civ. Dei 22.8, discussed below at n. 88.
Third century
The coexistence of multiple labor strategies in the North African countryside
and the development of tenancy arrangements that simultaneously balanced
the interests of laborers, lessees, and landowners created an economic
boom in the Maghreb that is well attested archaeologically. Interestingly, it
is precisely the sorts of crops most encouraged by the Lex Manciana that
enjoyed the most dramatic growth, especially the olive. Mattingly has
argued that by the third century, the province of Byzacena, and
particularly the semiarid Sahel inland from Hadrumetum as far upland as
Thelepte (Feriana), must have had a minimum of 10 million olive trees
which would have produced some 40,000 metric tons of oil per year by the
fourth century.44 The intensive field survey of the western Sahel around
ancient Sbeitla (Suffetula) and Kasserine (Cillium) has been able to
pinpoint the period of greatest expansion of oleiculture in this westerly
region to the third century, when stone-construction production facilities
with purpose-built oil presses were first installed in great numbers.45 Olive
cultivation in this region thus witnessed a massive growth that must have
been encouraged by the liberal land-tenure policies promoted on imperial
and probably also private estates through the spread of tenancy
arrangements based on the lex Manciana.
The economic and cultural power exerted by the agricultural and
demographic growth of producers has been emphasized in a number of
recent studies, especially in Dossey’s monograph Peasant and Empire in
Christian North Africa. Leslie Dossey argues that the fourth century
evidence for a restive peasantry is not reflexive of resistance to oppression
(as had previously been assumed) but rather of a newfound independence
fostered by economic prosperity and growing self–reliance.46 The same
spirit can be found in the passage of Agennius Urbicus quoted above
44
Mattingly 1988. Matttingly 1985 argues for similarly high figures (18,000 metric
tons / year) for Tripolitanian oil production by the third century. Mattingly, Stone,
Stirling and Ben Lazreg 2001 shows how the coastal city of Leptiminus became a
trading center, manufacturing goods for the export of oil and garum (amphorae) as
well as for agricultural use inland in the Sahel (metalwares). The regional
economic boom, which continued deep into the sixth century, was thus an urban as
well as a rural phenomenon.
45
Addyman 1962, 60–77; Hitchner 1988. Mattingly and Hitchner 1995, 195 argue
that some sort of tenancy arrangement, probably based on the Lex Manciana,
factored into the growth of North African oleiculture in this region at this time.
46
Dossey 2010, passim, esp. 193–203.
128 Chapter Six
which asserts that the tenants of private estates were often cohesive
enough to be regarded as a “plebeian populace” (populum plebeium) unto
themselves.47 The establishment of political solidarity around common
adherence to a large private estate is well attested in the epigraphic record
as well. One thinks for example of the vicani vici Annaei, who dedicated a
structure near the Municipium Semta (Ksour-Dzemda) in Proconsularis
using money provided them by Q. Geminius Arnensis Sabinus, a highly
decorated centurion under Trajan and, presumably, the owner of their
estate.48 Epigraphic examples can be multiplied. On the island of Zambrah,
near Carthage, the plebs fundi [3]itani dedicated an enclosure to the
goddess Ceres, again attesting to estate tenants claiming quasi-civic
status.49 The “Vesatenses”, who made a dedication in the 220s on their
estate at Bouraoui Belhadef (in Proconsularis) to their landlord, the Roman
senator C. Annius Anullinus Geminus Percennianus, show a similar sense
of group identity.50 Another dedication to Anullinus Percennianus was
made at Sidi Bou Skikine by a manager (actor) named Maximus, surely a
slave.51 This combination of slave manager and free tenants would seem
to offer an early example of a phenomenon much better attested in the
fourth through sixth centuries, when slaves were frequently used in high-
level managerial positions to supervise farms populated by coloni.52 But
perhaps the best example of this sort of “synoecism” of farm tenants in
North Africa comes in the Civitas Faustianensis in Byzacena, where the
tenant population of a private estate owned by a Q. Anicius Faustus
succeeded in achieving the status of a fully independent city in the early
fourth century.53 The strength of tenant communities is thus well attested
not just at an economic but also at a political level.
The archaeological record only confirms an impression of ongoing and
even expanding economic prosperity in this third–century period, very
47
Above n. 12.
48
ILTun 778–779.
49
CIL 8: 23022.
50
CIL 8: 27953 = ILAlg 1: 3636; cf. PIR2 A 633. Anullinus Percennianus is
probably the father or grandfather of PLRE I C. Annius Anullinus 3, who is
regularly attested as Proconsul Africae in Christian martyr acts of the Great
Persecution.
51
CIL 8: 27943 = ILAlg 1: 3625.
52
See Vera 1992.
53
M'Charek 2003 with Lepelley 1994.
much in contrast with traditional models of third–century collapse.54

Peyras has surveyed the North Tunisian farmstead at Bou Assid, known
from a dedicatory inscription to have been called the fundus Aufidianus
(or, by its local name, Biha Belta), and there found ongoing vitality under
labor and managerial relations like those seen in the second–century
inscriptions from the Bagrada Valley. The estate, which comprised almost
6,400 iugera (1,600 ha), was located in a region that received 550-600 mm
of annual rainfall and was thus rich in arable land. It was broken into
fifteen habitations: twelve small farms and three larger hamlets. The farms
appear to have averaged c. 200 iugera (50ha), about ten times the amount
of land necessary to sustain an average family with enough food to survive
year to year while paying rent and taxes. The inscription was dedicated by
the wife of a conductor whose name is now lost but whose achievements
included using grafts to bring a number of sterile wild olive trees into
production, digging a well, establishing an orchard, and planting new
vines.55 This was, in other words, precisely the sort of entrepreneurial
activity encouraged by the lex Manciana, undertaken on what appears to
have been a private estate, albeit by a conductor rather than his coloni.
Similar success stories can be told from the epitaphs of other North
African coloni, an indication that the growth in prosperity so evident in the
archaeology reached all the way down to the level of tenant farmers.56
This evidence, combined with what we can know about the size of
individual tenancies, paints a picture of tremendous prosperity and
profitability in what has usually been regarded as a period of economic
crisis.
A further indication of the ongoing preference for free tenancy in the
region can be found in testimonia for the North African uprising that
eventually vaulted the Gordiani onto the throne. When Maximinus Thrax’s
overzealous procurator of imperial estates in Africa attempted to extort
excessive tax and debt payments from the Carthaginian nobility, these
mustered their coloni from the countryside to form a peasant army,
overthrow the procurator, and eventually proclaim Marcus Antonius
Gordianus emperor at Thysdrus. Writing shortly after these events,
54
See Witschel 1999, 285–306 on the third century as a period of economic
flourish in North Africa.
55
AE 1975.883 with Peyras 1975; 1983.
56
See the verse epitaph of Locustius colonicus from Mactaris, CIL 8: 23427 = CLE
1870; cf. ILTun 243, the epitaph of Dion, a Christian, who lived 80 years and
boasted of having planted 4,000 olive trees––although he is not certain to have
been a colonus.
130 Chapter Six
Herodian describes the rebel band as consisting of “those cultivating the

land” (τὴν γῆν γεωργοῦντας) and “the mass of domestics” (τὸ μὲν πλῆθος
τῶν οἰκετῶν), by which he seems to mean the tenants of the large estates
whose owners coordinated the rebellion. This was certainly the
understanding of the late fourth-century author of the Historia Augusta
who reports that the rebel army consisted of “farmers or Africans”
(rusticos vel Afros) and was “a plebs both urban and rural” (apud plebem
vel urbanam vel rusticanam).57 This was not, in other words, a slave revolt
but a peasant uprising, coordinated by Carthaginian landlords using their
tenants to muster an army, for, as Herodian reports “Africa is by nature a
heavily populated country with many farmworkers on the land.”58 The
same preference for free tenants on private estates is also hinted at in a
letter of Cyprian indicating that he conceived of the dependents of
powerful North African patresfamilias as tenants (inquilini et coloni)
rather than slaves.59 This is not to deny that Cyprian’s extensive corpus
presents evidence for the ongoing presence of chattel slaves in Proconsularis,
but nowhere does he attest to their use in agriculture.60
Finally, the power of coloni to thrive and assert their claims to
authority over against the middlemen under whose authority they operated
on imperial estates is attested by a rescript of Constantine posted in
Carthage on March 9, 319 – CJ 11.63.1. This law is best grouped with the
third-century evidence because it predates Constantine’s law of 332
(discussed in the next section) which first attests to the binding of coloni to
their natal soil. As such it would seem to represent a continuation of older
traditions as yet unaltered by the harsher normative regime that would
shift the dynamic between landlord and tenant in the course of the fourth
century. CJ 11.63.1 was directed to emphyteuticarii, holders of life-long
leases on imperial land that were themselves heritable. By the late third or
early fourth century, this sort of very–long–term imperial lease had come
to replace the five–year leases that governed the conductores of the second
century, and it left a new breed of middle–men in a position of quasi–
ownership akin to that of the coloni themselves. The emphyteutic
57
Herod. 7.4.2–5; cf. Ioh. Ant. Hist. frg. 224 (Roberto) and SHA Gord. 7.2–4.
58
Herod. 7.4.4: φύσει γὰρ πολυάνθρωπος οὖσα ἡ Λιβύη πολλοὺς εἶχε τοὺς τὴν γῆν
γεωργοῦντας.
59
Cypr. Ep. 55.13.2 (CCSL 3B.270–71): ille qui inquilinos uel amicos suos ad
facinus conpulit et qui inquilinis et colonis pepercit.
60
For chattel slaves in Cyprian see Ad Demetrianum 8 (CCSL 3A.39); De bono
patientiae 6 (CCSL 3A.121); De lapsis 25 (CCSL 3A.234); De mortalitate 16
(CCSL 3A.25); Ep. 11.1.2–3 (CCSL 3B.57).
relationship might well have strengthened the position of these

leaseholders vis à vis the coloni by guaranteeing their rights to extended
control of imperial estates, yet the rescript itself indicates that the tenants
were very much still holding their own:
Some coloni have been doing harm to emphyteuticarii insofar as, contrary
to custom (praeter consuetudinem), they take over lands which they have
not developed by bringing into cultivation, although the tradition
(sollemnitas) has permitted these to work whatever has been planted up
either in olive trees or vines by their own labor. But they are also
attempting to use the flowing water from springs whose output is credited
to the emphyteuticarii alone. Therefore, we have decided that from now on
the right and power over these waters should remain with the
emphyteuticarii and that only so much may be used from it by the coloni as
is shown to be necessary for the farming of their own lands, which they
themselves cultivate. But insofar as there is surplus flow, which they might
use above and beyond their own crops, they should offer shares and access
points to it to the emphyteutic possessors.61
Several impressions emerge from this law. First, the provision of the
lex Manciana permitting coloni to gain quasi–ownership rights in the vines
and olives which they themselves had planted remained in force into the
early fourth century. Second, coloni felt sufficiently empowered in their
leaseholds to challenge the emphyteutic leaseholders, who technically
oversaw them, for control of arable land and water resources that should
rightly have been reserved to the emphyteuticarii.62 Third, still as late as
319, the emperor continued to benefit from the rivalry he had been
fostering since at least the early second century between his lessees
(conductores, and later emphyteuticarii) and the tenants they oversaw (the
coloni). By setting himself up as the mediator between his tenant–
managers and tenant-farmers, he succeeded once again in encouraging
both groups to compete with one another in devising strategies to
maximize their personal profits and in the process maximized his own
returns.
61
CJ 11.63.1 (my translation). In my analysis, I follow Vera 1988, 967–71; cf.
Courtois et al. 1952, 114–16; Wessel 2003, 111-13.
62
Vera 1988, 978–79 is surely right to emphasize that the lifetime leases of the
emphyteuticarii put them in a much stronger negotiating position vis à vis their
tenants than the second–century conductores, who generally took five–year leases.
Even so, both sorts of long–term leaseholders found their tenants to be a serious
match for their authority.
132 Chapter Six
Fourth century
Of course, Constantine’s rescript was issued shortly before his famous law
of 332 (CTh 5.17.1) that represents our first firm evidence of what has
been termed the “bound colonate.” This is a subject of profound
importance over which disputes have raged for centuries. At present
scholarly opinion is starkly divided over whether “the colonate” in Late
Antiquity represented a definable juridical status or simply a fiscal
category meant to register tenants on tax rolls that has been overblown into
an imagined quasi-serfdom, little more than a “mythe historiographique.”
Without being able to enter into the details of this still unresolved debate
in this study, I state up front that I hold with more traditionalist
interpretations that regard the ius colonatus as a new juridical status that,
although it was developed only over the course of a century and although
it could never be implemented uniformly in a world that was crumbling
into a variety of political jurisdictions even as it came into being,
nevertheless represented a de facto third legal class of persons halfway
between free and slave.63
Regardless of where one stands on the question, Constantine’s law of
332 remains important for this investigation as a watershed in relations
between tenants and landowners. Whether or not the imperial government
was effectively creating a new personal status, it demonstrably was
attempting to limit the personal mobility of coloni to the estate on which
they were born, and this would begin to have profound consequences on
the personal freedom and economic potential of late antique tenants.64 Of
itself the effort to hold tenants on the land of their origo represented an
important shift from earlier patterns, for where attachment to landed
estates had previously been encouraged through the grant of perpetual usus
proprius over any newly developed marginal land, there had been no
requirement that coloni or their offspring remain on the land of their birth
63
For the traditionalist view, see Vera 1987; 1992–1993; Mirkovic 1997; Banaji
1997; Giliberti 1999; Sirks 2008. Kehoe 2008, 163–91 takes a middle ground,
accepting the overwhelming weight of evidence for legal restrictions on movement
by coloni and emphasizing its negative impact on agricultural performance, but not
treating the colonate as an effective new status. For the argument that the colonate
is a mythe historiographique, see Carrié 1982; 1983; 1997; Grey 2007; 2011, also
discussed above at n. 5.
64
We should not lose sight of the fact that there are indications that some aspects
of this policy appear already to have been in place as early as 319, cf. CTh 11.7.2
(Nov. 20, 319); CJ 11.68.1 (Oct. 7, 325).
(origo) in perpetuity. Naturally, the aspirations of the state to lock tenants

to their origo often faltered in the face of real–world complexities. The
source record makes it clear that both the state and private landowners
were regularly stymied in their efforts to restrict the mobility as well as the
personal, social, and economic freedom of coloni.65 In response to these
complications, however, the imperial government did not abandon its
efforts to build out a failsafe legal framework for the restriction of coloni
but rather redoubled them, refining the principle of binding tenants in at
least three ways:
• First, it responded to problems that arose from attempts to

circumvent the binding of coloni to their estate of origin by
forbidding the sale or transfer of land without its coloni (CTh
13.10.3 = CJ 11.48.2 [a. 357]) and by forbidding the division of an
estate in order to redistribute its coloni in ways that effectively
detached them from their origo (CJ 11.48.7 [a. 367/75]: cf. CTh
11.1.26 [Jun. 19, 399]).
• Second, it gradually took steps throughout the course of the fourth
century to extend the principle of binding coloni into those
provinces where this had not yet been the practice (Illyricum = CJ
11.53.1 [Jun. 29, 371]; Palestine = CJ 11.51.1, cf. CTh 5.17.2 = CJ
11.64.2 [Oct. 25, 386]).
• Third, it began assimilating the status of bound tenants to that of
slaves. In some ways this move was a linguistic sleight-of-hand, as
seems to be most clearly attested in a famous law of 393 (CJ
11.52.1; cf. CTh 13.11.4) which proclaims that coloni shall be
bound by the law of their origin (originario iure) so that, although
they appear to be of free condition, “they shall be considered as
slaves of the soil on which they were born” (servi terrae ipsius cui
nati sunt aestimentur). While it has been argued that this statement
is more a metaphorical analogy than a legal principle, the logical
slippage it represents had practical effects. The assimilation of
bound coloni to slaves had quite real consequences for the tenant
which included at least seven impediments to personal freedom not
imposed on other free persons:
1. Marriages between freedmen (liberti/ae) and imperial coloni/ae
were restricted (CJ 6.4.2 [a. 367])
65
A fact emphasized throughout the fine study of Grey 2011.
134 Chapter Six
2. The status of colonus became heritable across generations (CJ

11.52.1 [a. 371]; cf. CJ 11.48.13; CTh 5.18.1; Nov. Val. 31.5-6)
3. Any property acquired and owned by a colonus was treated as a
precarious grant of his or her landlord under the title of
peculium (slave property) (CTh 5.19.1 [a. 365]; cf. CTh 5.19.2;
5.18.1)
4. Landlords were permitted to distrain upon the person of their
colonus (CTh 5.17.1 [a. 332]; cf. CJ 11.53.1; CTh 14.18.1;
16.5.52.4; 16.5.54.8)
5. Like fugitive slaves, coloni who fled their home estate were to
be captured and forcibly returned (CTh 5.17.1[a. 332]; CJ
11.48.8; 11.53.1; CTh 5.17.2–3).
6. Criminal or undesirable behavior on the part of free men could
result in subjection to the colonate as a judicial penalty—much
like enslavement for crimes (servitus poenae) (CTh 14.18.1 [a.
382])
7. Mechanisms were created for the liberation of coloni from their
status based in late antique provisions allowing for the
acquisition of freedom by slaves through long-term prescription
(CTh 5.18.1 [a. 419]; cf. Nov. Val. 27; 31; 35.18–19; CJ
11.48.16).
Thus, regardless of where one stands on the question of whether the

agglomeration of impediments imposed on late Roman coloni amounted to
a discreet and coherently defined new personal status, there can be no
doubt that the social and economic power of coloni came to be severely
impaired by the imposition of new state regulations.
We have very little evidence with which to pinpoint the status or even
just the activities of coloni in North Africa during the fourth century, when
this shift in the status of farm tenants was occurring. We also have almost
nothing with which to determine the relative importance of slaves versus
coloni in the North African agricultural labor pool during this century.
Thus, while the mid-fourth century Expositio totius mundi claims that
Mauretania was a net exporter of slaves, this report says nothing of
whether these slaves derived from local estates (which would likely
indicate the widespread use of agricultural slavery) or from captive
Africans, whether Berber or sub–Saharan. Given that the captivity and sale
of black Africans is well documented from at least the second century
onward, this latter is the most likely explanation, which makes the
Expositio useless as a source for agricultural slavery in the Maghreb.66 In a

famous passage of Optatus describing unrest connected with the Donatist
controversy in the mid-340s, we learn that in central Numidia in the mid–
340s during the uprising of Axido and Fasir masters were afraid to travel
public roads since they feared being set upon by their slaves, who seized
their carriages and made them run alongside in servile fashion. The
passage does not clarify, however, whether these were domestic or
agricultural slaves, and a strong argument has even been made that
Optatus is most likely describing debt slavery among impoverished tenants
rather than chattel slavery as such.67 Finally, the only letter of Symmachus
firmly attesting to his ownership of an estate in Mauretania Caesariensis
says nothing of its labor force. If, however, the evidence from
Symmachus’s Italian estates is any indicator of his preferences regarding
the organization of his agricultural workforce, these are likely to have been
coloni rather than slaves.68
Apart from the law of Constantine of 319 mentioned above (CJ
11.63.1), a law that precedes the first solid evidence for the bound
colonate by some thirteen years, we have only one fourth-century
normative regulation relating to farm tenancy in North Africa specifically:
a fragment of a constitution of Valentinian I issued in 366 to the governor
of Tripolitania which orders estate holders (domini) to accept tax payments
from their peasants (rustici) in kind and not demand cash (CJ 11.48.5). We
do, course, have an abundance of laws on the colonate (only a sampling of
them is listed above), but most cannot be linked to any specific region.
One factor of interest here is that the assimilation of coloni to slaves
attested more broadly in the normative sources appears to have played
itself out at the ground level by making both types of laborer seem
66
Expositio totius mundi 60 (SCh 124.200). For captive African slaves, see CIL 8:
21486 = ILS 4495; CIL 8: 4508 = 18643; Aug. Ep. 199.46 (CSEL 57.284–5).
67
Optat. 3.4 (CSEL 26.82): itinera non poterant esse tutissima, quod domini de
vehiculis suis excussi ante mancipia sua dominorum locis sedentia serviliter
cucurrerunt. Illorum iudicio et imperio inter dominos et servos condicio
mutabatur. See Shaw 2011, 781–2: “The whole context, which speaks of nothing
else, excites the strongest suspicion that these slaves were not chattel slaves
(persons owned as property) as is normally thought, but rather debt–slaves who
were seeking freedom from the unjust treatment imposed on them by the terms of
debt–bondage.... it is most probable that the documents, the tabulae, that were
being destroyed were records of debts owed by peasant workers to landlords.”
68
Symmachus Ep. 7.66. Vera 1986, 258–59 shows beyond doubt that
Symmachus’s Italian labor force consisted largely of coloni.
136 Chapter Six
interchangeable on imperial estates. Thus a law of 383 addressed to an

eastern Praetorian Prefect forbids older coloni from being kicked off of
their imperial tenancies and replaced with slaves or new coloni.69 Thus the
emperor in the fourth century continued to intervene in support of the
rights of his tenant farmers, much as he had done in the second and third
centuries. Ironically, however, in this instance, it was imperial policy itself
which had created the circumstances that weakened the position of tenants
vis à vis their middleman managers by limiting their freedom of
movement and assimilating them ever more to the slaves by whom they
were at times replaced. The law cautions against any monolithic
assumptions about the exclusive use of coloni rather than slaves to farm
imperial estates.
Overall, then, the meager harvest of materials for the fourth century
rural labor pool in North Africa indicates that not only were slaves by no
means the only source of rural labor in the Maghreb, they were likely to
have been considerably outnumbered by tenants.70
Early fifth century

The effects of imperial policies binding coloni to the estates of their origo
in North African agriculture become fully apparent in the textual and
documentary sources for the early fifth century. Both source pools reveal
the outcome of three quarters of a century’s worth of laws that gradually
subordinated a formerly self-assertive and surprisingly independent body
of tenant farmers to slaves. On the normative side, this shift is clearest in a
set of three laws connected with the Donatist controversy. Beginning in
405 the emperor Honorius began to take a decidedly harsher stance against
adherents of this schismatic sect. At the heart of his discomfort with
Donatists was their (heretical) sacramental praxis of re-baptizing new
adherents. In his zeal to occlude this abomination, Honorius took
advantage of the subordination characteristic of fifth–century coloni to
insist that any estate owners who forced “slaves or personnel subject to
their power” (servos vel homines iuri proprio subditos) into rebaptism
were to have their estates confiscated, and any managers (conductores and
procuratores) who did so without the owner’s knowledge were to be
69
CJ 11.63.3.
70
So also Vera 1987. On the empire as a whole, see Vera 2012, aimed directly at
Harper’s broader thesis. On Italy, see already Vera 1992-1993; cf. Wickham 2005,
259-302.
beaten with lead–tipped whips (CTh 16.6.4). The law not only confirms
the ongoing use of slaves alongside agricultural tenants in this period but
also points to the degree to which the slave/colonus equivalency had
turned coloni into pawns of their masters and managers in the larger
sphere of religious politics. The two remaining laws on fifth–century
African coloni date to 412 and 414 respectively and were introduced in
order to enforce the findings of the Council of Carthage in 411 at which it
was ordered that all Donatists must reunite with the Catholic church. An
elaborate series of fines was established graded according to rank; on this
scale, slaves and coloni were set on a par and ordered to be converted
either through admonitions or, when necessary, regular whippings by their
masters.71 The 414 law changed tack only slightly by ordering smaller but
more frequent fines for free men who persisted in their Donatism, but
continued to insist on beatings for slaves and coloni and confiscations for
those latter who refused to cooperate.72 Here again, the laws indicate that
slaves and coloni worked alongside one another on rural estates; that the
two were considered rough equivalents; and that as such both were subject
to harsh corporal punishment at the hands of landowners and/or managers.
The corpus of Augustine offers confirmation of the manner in which
the fifth-century landowner was enlisted as a vehicle to carry out the
religious agenda of the Roman church and its secular advocate, the
emperor. The circumstantial details Augustine provides also help fill out
our understanding of the functioning of land tenure and rural labor
organization in the period. His Letter 58 of 401 was written to the Roman
senator Pammachius to congratulate him on enforcing the adherence of his
coloni to Catholicism. Not only does this inform us of the collusion
between landlord and bishop in the enforcement of Catholic orthodoxy
among tenants, it also reconfirms the involvement of Italian elites in North
African agriculture and points to a preference for tenant laborers (coloni)
over slaves on the part of Pammachius.73 A letter of c. 400 to Crispinus,
71
CTh 16.5.52, esp. 4: Servos etiam dominorum admonitio vel colonos verberum
crebrior ictus a prava religione revocabit, ni malunt ipsi ad praedicta dispendia,
etiam si sunt catholici, retineri.
72
CTh 16.5.54, esp. 8: Servos vero et colonos cohercitio ab huiusmodi ausibus
severissima vindicabit. Ac si coloni verberibus coacti in proposito perduraverint,
tunc tertia peculii sui parte multentur. More on these laws at Vera 1992, 465-67.
73
Aug. Ep. 58.1 (CSEL 34.217): ...non tibi tam dilecta catholica unitas foret nec
colonos tuos Afros eo terrarum, unde donatistarum furor exortus est, hoc est in
media Consulari Numidia constitutos tali admoneres adloquio. See also Aug. Ep.
112.3 (CSEL 34.659), where Augustine encourages another senatorial landlord
138 Chapter Six
the Donatist bishop of Calama, presents the same case from the other side.
In it Augustine upbraids his addressee for re-baptizing the farm laborers
on an estate called Mappalia over which he had recently acquired
possession. Augustine provides further information about the same issue
in his treatise Against the Letter of Petilian which reports that what
Crispinus had actually purchased was not the estate itself but rather the
emphyteutic rights to an imperial estate populated by c. 80 coloni. From
the combination of the two sources a picture emerges of a Donatist
bishop’s attempts to enforce his own version of Christian religion on an
estate he controlled––one technically owned by a Catholic emperor––
through the forced rebaptism of its labor force. The letters also make it
clear that said labor force consisted entirely of coloni.74 Crispinus’s
Mappalia was thus on the same scale as the third century fundus
Aufidianus mentioned above with its fifteen habitations each of which
could support a farmer and his family of approximately five individuals.75
Land tenure and labor organization had thus changed very little between
the first and fifth century, even if the status and privileges of the tenant
farmers who cultivated it most certainly had.
Augustine’s letters also confirm that the manipulation of late antique
coloni did not stop with religious politics. Good old–fashioned fraud and
chicanery were also abundantly on show. A striking example of this can be
found in letter 247 to a certain Romulus, a landholder in the territory of
Hippo Regius, whose agent (actor) had collected twice the rent he was
actually owed from his coloni. Augustine asked that Romulus pity these
“wretched and impoverished people” (miseri et egeni homines) and return
what they did not owe, but his plea appears to have fallen on deaf ears.76
Augustine’s New Letter 10* reconfirms the level of poverty to which
coloni were often reduced when it describes a colonus who felt the need to
named Donatus, the former proconsul Africae, to enforce the conversion to

Catholicism of his dependents (tuos omnes)—presumably coloni.
74
Aug. Ep. 66.1–2 (CSEL 34.235) with C. litt. Petil. 2.83[184] (CSEL 52.114).
75
Compare Gregory the Great’s effort to manipulate pagan tenants on the
ecclesiastical estates in Sardinia to convert by arbitrarily increasing their rent until
they agreed to comply, Greg. Mag. Ep. 4.26 (CCSL 140.244). Gregory used
similar tricks in Sicily, where he ordered his deacon Cyprian to offer reduced rents
to those Manichaeans and Jews who agreed to convert, Greg. Mag. Ep. 5.7 (CCSL
140.273).
76
Aug. Ep. 247.1–3 (CSEL 57.586–8); cf. PCBE I Romulus 2. On this letter, see
Cracco Ruggini 1987.
sell his own wife to slave dealers.77 Similarly, New Letter 24* poses a
series of questions about the legal implications of a law of Constantine’s
permitting fathers to sell their children into slavery, or rather long term
servitude78: if a colonus did so, did this trump the claims of the landlord to
the child’s labor? Could female colonae also sell their children? Could
landlords sell the children of their coloni? Of course, all of this confirms
the co-existence of slaves alongside coloni, but Augustine’s concerns in
the letter seem to be that the sale of colonus children would extract them
from the rural labor force of their landlord altogether. Indeed, the whole
import of the New Letter 10* is that many of Hippo’s peasants had been
precipitously captured or bought by Galatian slave traders (mangones) for
export out of Africa. African coloni were not, then, being recycled to serve
as slaves on other local estates. Rather an African labor force consisting
primarily of coloni was being enslaved for resale in other regions.
A slightly more sanguine picture emerges from New Letter 20* which
describes how the coloni on the fundus Thogonoetensis, yet another
African estate under the ownership of an absentee senatorial landholder,
exercised at least some power over their own destiny by threatening to flee
en masse if they were assigned the infamous swindler Antoninus of
Fussala as their bishop.79 Here again, we have confirmation that tenancy
remained the primary mode for organizing agricultural labor on large
estates in North Africa, and we catch at least a glimpse of the same self-
assertiveness we witnessed in tenants of an earlier age.
Even so, as New Letters 10* and 24* indicate, the line between slave
and colonus had become quite blurred. There is confirmation for the same
from a letter composed by a middling aristocrat from the town of Matar
(modern Mateur) in Proconsularis and preserved in manuscripts of
Orosius.80 This text reveals that its anonymous author had an ongoing
dispute with a certain Salvius, an advocate, who laid claim to all or some
of the coloni on the author’s estate, the fundus Volusianus. These lived in
mortal fear of being separated from their natal soil in a pending court case,
leading the letter writer to plead with Salvius to settle the dispute
77
Aug. Ep. 10*.6 (CSEL 88.46–51). See also Cass. Var. 8.33 (CCSL 96.340–1)
which describes a rural fair in Lucania where rusticani were known to have sold
their children into slavery.
78
Aug. Ep. 24*.1 (CSEL 88.126) with Lepelley 1983. Constantine first permits
child sale with CTh 5.10.1 and CJ 4.43.2 (a. 329).
79
Aug. Ep. 20*.9–21 (CSEL 88.100–106). See also Ep. 20*.29 (CSEL 88.110)
which confirms that Antoninus also used rental tenancy for the land he managed.
80
Lepelley 1989.
140 Chapter Six
amicably. Here we have yet another detailed source confirming the picture
supplied by Augustine of a fundamental reliance on a tenant labor force
that was, nevertheless, so tightly bound to their origo that they could be
forcibly relocated as the result of a judicial ruling.
Finally, as if we needed further evidence of the same pattern, in his
description of the Donatist Crispinus of Calama’s attack on his rival
Catholic bishop Possidius at the fundus Olivetensis, Augustine makes it
clear that this estate was also populated with coloni.81 So too, in his
discussions of the ecclesiastical estate of Hippo he describes at Ep. 35 and
of a Numidian farm called the fundus Strabonianensis mentioned at Ep.
65, Augustine mentions only free tenant farmers.82
When we arrive at the detailed information for rural labor management
provided by the corpus of Augustine and other early fifth century sources,
the conclusion is thus inescapable that the prevailing form of labor
organization was tenancy. Fundamentally, then, coloni rather than slaves
worked North African agricultural estates in the age of Augustine. To be
sure, when subsistence crises or cash scarcity necessitated, the abundant
supply of coloni could be capitalized as human chattel––as the New
Letters reveal––but the demographic basis of rural labor consisted of
bound tenants.
To say this is not to deny that slaves were still used in fifth–century
African agriculture. In the laws on Donatism described above, we have
seen that the emperor and his agents conceived of slaves and coloni as
equally probable laborers on North African agricultural estates.83 In
several narrative passages, Augustine himself also assumes that both
slaves and coloni could be used to cultivate farmsteads.84 Thus, in a
sermon defending two of his newly appointed sub-deacons for having
heretofore failed to liquidate their landed property despite ordination,
Augustine mentions that both had slaves (mancipia) who, we can assume,
were used to cultivate that property. Indeed, Jerzy Kolendo has argued that
small to medium-sized landholders regularly used their moderate familiae
81
Aug. Ep. 105.4 (CSEL 34.598): nisi tertio suppositas flammas coloni eiusdem
fundi propter periculum suae salutis extinguerent; cf. Cresc. 3.46[50] (CSEL
52.458).
82
Aug. Ep. 35.4 (CSEL 34.30–31). For the fundus Strabonianensis, see Ep. 65.1
(CSEL 34.323–33).
83
See above nn. 71-73.
84
See E.g. Aug. En. in Ps. 34(1).12 (CCSL 38.308): si possessionum es amator,
desideraturus es totam terram, ut omnes qui nascuntur, coloni tui aut serui tui sint;
Aug. C. Adim. 15 (CSEL 25.159); En. in Ps. 80.9 (CCSL 39.1125).
for cultivation in addition to the variety of domestic tasks assigned them.85

There are also the oft repeated passages from the Latin Life of Melania the
Younger reporting how this super–rich heiress and her blue–blooded
husband Pinianus sought to liquidate their worldly wealth, including
estates in Numidia, Mauretania, and Africa Proconsularis, along with their
slaves. The astronomically high number of slaves the early fifth-century
writer Palladius claims Melania manumitted (8,000) does not, however,
appear to have come substantially from her African estates. Instead, the
author of her Latin Vita reports, she and Pinianus settled the slaves they
freed from North African estates in a monastery they founded in Thagaste,
where they numbered 130 women and 80 men. If these figures represent
the total from their slaveholdings in the region, as seems to be the author’s
implication, this would have represented a relatively small workforce –
enough to populate two latifundia, or perhaps more likely, the domestic
staff from a larger number of estates whose primary agricultural workforces
consisted of coloni.86
The overall impression we get from the flood of early fifth-century
material, and particularly from Augustine’s extensive and circumstantially
detailed corpus, is that coloni were considerably more important than
slaves as agricultural workers by this point in the history of the region. In
fact, in every instance but one in which Augustine describes in any detail
the labor force of an actual early fifth-century North African estate,
agricultural manpower appears to have consisted entirely of coloni.87 To
be sure, Augustine takes great delight in discussing slaves and slavery in
his many exegetical descriptions of scripture, and these often build on
metaphors and parables involving agricultural slavery that had been
knitted into the fabric of Christian scripture and exegesis since the
85
Aug. Serm. 356.3 (PL 39.1575–76) with Kolendo 1991, 43; cf. Génelle 2005,
426–29.
86
Vita Melaniae Latina 20-22 (Laurence pp. 190-96); cf. Gerontius Vita Melaniae
Graeca 20-22 (SCh 90.168-72). For the report of 8,000 manumissions, see Pall.
Hist. Laus. 61 (Butler 156).
87
Aug. Civ. Dei 22.8 speaks of a fundus Zubedi that appears to have been
populated with both slaves and tenants. It is not clear, however, that slaves worked
the fields there rather than performing domestic services. In fact, En. In Ps. 79.10
(CCSL 39.1116) would seem to imply that a slave was inclined to function as an
estate messenger while coloni did the agricultural work. Harper 2011, 184 quotes
En. In Ps. 103(3).9 (CCSL 40.1506–7) as an example of agricultural slavery, but
the passage seems rather to treat household slavery in an urban dwelling (domus).
142 Chapter Six
composition of the Gospels.88 But in descriptions focusing on the

circumstantial details of real North African estates—the nuts and bolts
cases with which Augustine regularly dealt in his pastoral and juridical
capacities—tenancy appears to have been the predominant model, albeit
the tenancy of coloni whose status had been reduced to little more than
slavery. This is important because much of Harper’s argument for the
ongoing survival of Rome’s “slave society” in the late antique West is
built on evidence from the corpus of Augustine. However, not only is
direct testimony of the use of slaves in agriculture relatively thin in
Augustine, the evidence for the primary reliance on coloni as farm workers
is frankly overwhelming. It is only Harper’s choice not to examine this
evidence that has led him to argue otherwise.
Vandalic period
The Vandalic takeover of North Africa in the 430s by all means ushered in
a shift in political structures, but the question has long remained, did this
have any measureable effect on economic and social relations. Current
scholarly opinion tends to emphasize continuities with the Roman past,
and these are certainly evident in the sources.89 There is some indication,
however, that at least in terms of land tenure and labor management,
significant changes occurred. At a minimum, there is almost no way for
scholars to deny the ample testimony that the Vandals expropriated
property from a large number of elite land–holders.90 Moreover, there is
abundant evidence that the Vandals enslaved many Roman residents of
North Africa in the period of their conquest. To be sure, most of our
sources on the matter point to the enslavement of members of the elite, but
there are also indications that captive taking and enslavement occurred on
a more generalized level.91 In the years following, enslavement was used
88
For a full list of Augustine’s repeated references to slaves and slavery, the vast
majority metaphorical, see Klein 1988.
89
Merrills, and Miles 2010; Conant 2012; Modéran 2014.
90
Modéran 2002.
91
Captivity of elites: Vict. Vit. Hist. 1.13–14; Procop. Bell. 3.5.11; Prosp. Tiro
Chron. 1339 (s.a. 439) (MGH AA 9.477). General captivity: Nov. Val. 13.14;
Possidius Vita Augustini 28; 30.3–14; Quodvultdeus De tempore barbarico 2.1.1–
3, 5.2–23 (CCSL 60.473, 476–77); Salvian. Gub. 6.12.69–70 (CSEL 8.144–45).
Geiseric also enslaved captives seized after Aspar’s failed expedition against his
kingdom in 434, Procop. Bell. 3.4.2–3; Evag. Schol. HE 2.1 (Bidez and Parmentier
pp. 37–38).
by the Vandals as a standard form of punishment;92 slaves were also

imported through trade;93 and, above all the Vandals engaged in regular
raids on the coasts of Sicily, Lucania, Bruttium, Campania, Illyricum, the
Peloponnessus, and the Aegean islands, which netted not just plunder but
also large numbers of captives.94 Most notorious among their targets was
of course Rome, whose capture in 455 yielded not only massive amounts
of plunder but also “many thousands of captives” including the empress
Eudoxia.95 Victor of Vita records that these Roman captives were
transported by sea back to Carthage, where they proved to be so numerous
that two sizable basilicas were set aside as corrals to manage them until
they could be redistributed to slaveowners.96 Some of these wound up on
Mediterranean slave markets. We know, for example, of a western
aristocrat named Maria who was sold together with her handmaid to
merchants that eventually resold her in Syria.97 Other victims of Vandal
raiding expeditions, however, were likely deployed in North African
agriculture.
The evidence for the use of agricultural slaves by the Vandals is, like
the evidence for the Vandal kingdom more broadly, tenuous. Victor of
Vita does report that, as part of his general persecution of Nicenes,
Huneric enslaved Catholics working in his court and sent them to Utica to
labor in the fields.98 He is also said to have punished a certain Gamuth,
brother of Heldica, by sending him to the countryside to cut sod and plant
vines, probably as a slave.99 More tellingly, Procopius reports that in the
92
Vict. Vit. Hist. 1.43–46; 2.10–11; 2.15–16; cf. 3.62.
93
Fulg. Rusp. Ep. 11–12 (CCSL 91.360–63); cf. Procop. Bell. 3.10.25–34.
94
For raids reported explicitly as having netted captives, see Procop. Bell. 3.5.22–
23; Malchus Hist. fr. 5 (Blockley = Exc de Leg. Rom. 3); Gelasius Ep. 17 (Thiel
381); Priscus Exc. 30.3 (Carolla = Blockley 39.1 = Exc. De Leg. Gent. 14).
95
Prosp. Tiro Chron. 1375 (s.a. 455) (MGH.AA 9.484): multaque milia
captivorum, prout quique aut aetate aut arte placuerunt, cum regina et filiabus
eius Cartaginem abducta sunt; Victor Tun. Chron. 15 (CCSL 173A.7–8).
96
Vict. Vit. Hist. 1.24–27; cf. 1.30–38.
97
Theod. Ep. Sirm. 70 (SCh 98.152–4), datable to 443/448. During his persecution
of Nicenes, Huneric also traded away catholic monks and nuns as slaves to the
Mauri, Passio Beatissimorum martyrum qui apud Carthaginem passi sunt sub rege
Hunirico 2 (Lancel 2002, 213–14). He also sold Manichaean heretics as slaves in
order to buy ships, Vict. Vit. Hist. 2.1.
98
Vict. Vit. Hist. 2.10.
99
Vict. Vit. Hist. 2.15–16. Conant 2012: 102–3 interprets these as incidents of
mere exile, but the aspect of forced labor makes it clear that some form of penal
slavery was involved.
144 Chapter Six
second year after the East Roman reconquest of the Vandalic kingdom, a
revolt arose in large part because Roman soldiers stationed in North Africa
had married Vandalic women who were then asserting their claims over
the land their families had once farmed; this occasioned trouble when
Justinian’s general Solomon reported his willingness to grant them control
of these women’s money and slaves but not the land, which he ordered to
be surrendered to the emperor.100 The implication seems to be that the
farms were cultivated for their Vandal owners by slaves. In the passage
that follows, Procopius reports that the resultant uprising, led by the native
general Stotzas, was bolstered by the participation of many of these slaves
in the armies of their former masters.101 It would seem, then, that this
period of intensive slaving on the part of the Vandals led to an
intensification in the use of slave labor in regional agriculture. If so this
would be entirely in keeping with historical patterns––ancient and modern
––that have witnessed the intensification of slave labor in periods that saw
quantum leaps in captive taking.
Despite this turn to slave labor, the Vandals also surely maintained
tenancy as an established method for exploiting the estates which they had
taken over from their Roman predecessors. We have one piece of textual
evidence that confirms this, a passage in Victor of Vita reporting that
Huneric punished Catholic clergy who had attempted to placate him by
swearing an oath of loyalty––contrary to the Gospel prohibition against
swearing at Matt 5:13––by condemning them to serve as coloni in the
fields of North Africa. Interestingly, those who refused the oath were
punished with exile to Corsica, where they served as penal slaves cutting
timber.102 Huneric’s persecutions also led to the enactment of a decree to
punish procuratores and conductores on royal estates (regalia praedia)
with a doubling of their rent payments, an indication that farm
management had changed little on estates belonging to the crown, even if
we cannot confirm whether the laborers on these farmsteads were slave or
free.103
100
Procop. Bell. 4.14.8–10.
101
Procop. Bell. 4.15.3-4. Fulgent. Rusp. Serm. 1.1-3 (CCSL 91A.889-90) also
points to the use of agricultural slavery in Vandalic Africa, but in an exegetical
passage that cannot be firmly associated with a particular estate.
102
Vict. Vit. Hist. 3.19–20. The Corsican exiles appear to have numbered just 46 if
these are to be associated with the catalog in the Notitia Provinciarum et Civitatum
Africae, Lancel 2002, 252–72.
103
Vict. Vit. Hist. 3.11.
There are, however, clearer indications of the ongoing importance of

tenancy in the fascinating documents known as the Albertini Tablets,
discovered in 1928 not far from Tebessa in eastern Algeria. These record
purchase transactions from a variety of tenants on the fundus Tuletianos in
the 490s. One actually attests to the acquisition of a six–year–old slave
boy,104 but most record the sale of small parcels of land together with their
fruit trees––primarily olives, but also vines, pistachios, and almonds. One
document seems to confirm that the tenants involved continued to pay rent
shares.105 Interestingly, the tenants generally refer to themselves as citizens
(civis, or more commonly cibis) of the fundus Tuletianos, very much in
keeping with earlier testimonies to the assumption of a quasi-civic status
based on attachment to a farmstead that we witnessed in Agennius Urbicus
and various epigraphic testimonia. An even more striking point of
continuity can be found in the fact that in some 13 of the total of 34 tablets
the tenants refer to their holdings as “Mancian cultivations.”106 This
curious survival makes it clear that, well into the Vandalic period, tenants
not only continued to operate as producers on rural estates but also clung
to earlier Roman normative regulations that guaranteed their ownership
rights over the usus proprius of the plantations they or their ancestors
initiated.
The Vandals may then have intensified the employment of slaves in
their century of rule in North Africa, but tenancy clearly remained a major
method of organizing rural labor. Insofar as this is true, any model that
assumes the strength of the Roman economy was the primary driver of
intensification in the use of agricultural slavery must be revised. The
violence inherent in the Vandal takeover of Africa is well attested, and
despite Vandalic efforts to maintain political and economic continuities
with their Roman predecessors, it surely led to a temporary increase in the
percentage of slaves in the North African population. While it is unlikely
these overtook tenants as the primary producers of surplus for the elite, we
at least have stronger indications in this direction than for previous
104
Tab. Albertini 2 (Jun. 5, 494) at Courtois et al 1952, 216–17.
105
Tab. Albertini 26, at Courtois et al 1952, 288: ita placuit ut secundum quod est
in condi/tionem quod in polepticos clarit fici arbores quindecim annos quinque / et
olibe arbores quindecim ut exsudet pensionem s(oluat).
106
Ex culturis suis mancianis: Tab. Albertini 4, 9, 10, 11, 12, 13, 14, 17, 19, 20,
22, 23, 24. More on the tablets at Dossey 2010, 119–20; Conant 2012, 281–83,
with earlier bibliography. Already in the late second century we have attestations
to tenants making express claims to Mancian tenancy, ILTun 629 = AE 1938, 72:
C(aius) Aufidius Utilis Manciane cultor.
146 Chapter Six
centuries after one discounts for the paucity of information about Vandal
North Africa.
Sixth century
With the Roman reconquest of North Africa in 533–534, we witness an
effort on the part of the eastern court to reestablish the bound colonate as
the dominant mode of agricultural production in the region. One can
already get an inkling of this preference of the Roman government for the
structuring of rural labor around tenancy rather than slavery in two
fascinating documents from the reign of Valentinian III. Both were
designed to restore fiscal stability and repopulate imperial estates in the
provinces of Mauretania Sitifensis, Mauretania Caesariensis, and Numidia
Constantina after these had been retroceded to the imperial government by
the Vandals in a treaty of 442. In the first, Novel 13 of 445, Valentinian
radically cut tribute rates in these provinces by seven–eighths and offered
a variety of further indulgences to recalibrate tax and service burdens,
including strictly forbidding the ordination or military enlistment of
tenants because of the current “scarcity of coloni” (13.8: raritate
colonorum). More tellingly still, in his Novel 34 of 451, Valentinian
ordered:
And in the province of Sitifensis and Caesariensis, estates under

emphyteutic law and those belonging to the imperial fisc which had fallen
to neighboring dwellers after the destruction by the Vandals and are today
possessed by various people, I decree should be taken away from these
same and under that form of apportionment by which they are now held
(sub eo pensitationis modo, quo nunc tenentur), and with any imperial
rescripts being held in abeyance, they should be rented to honorati from
the provinces of Proconsularis and Byzacena, whom we know to have been
expelled from their own homes when their properties, which were
confiscated, were taken away by the barbarians.107
Valentinian was thus struggling to re-impose upon imperial holdings as

quickly as possible the age old system of long-term tenancies, presumably
under the lex Manciana. He believed he could implement this goal using
elite landholders who had been displaced from what had now become the
heart of the new Vandal kingdom in Proconsularis and Byzacena. In other
107
Nov. Val. 34.3 (a. 451); cf. Vict. Vit. Hist. 1.14 on the exile of bishops, clari
and honorati by the Vandals.
words, the western emperor wished to revive the system of elite

management of his imperial estates by transplanting conductores and
emphyteuticarii from Proconsularis to Mauretania where they could
continue the time-tested pattern of overseeing tenancies populated by
share-cropping coloni.
Much the same occurred in the aftermath of the Justinianic reconquest
of the sixth century. By this point the extent of the disruption to traditional
patterns created by a century of Vandalic rule in Proconsularis and
Byzacena was much clearer. This we learn from two Novels of Justinian,
the first of which is dated to 552 and addressed to the Praetorian Prefect of
Africa Paulus. Paulus is informed that those coloni who had escaped from
their status while under the Vandals were not to be recalled to their estates,
but that those who had merely moved from one estate to another were to
be recalled to their original holdings.108 A second law, dated to 558,
reaffirms the first principle, which landowners had been attempting to
circumvent, by restating that all those held as coloni who had gained their
freedom or been ordained clerics before the arrival of Justinian’s armies
were to remain in freedom.109 These provisions would seem to confirm the
theory elaborated above that the Vandals neglected the complex normative
rules governing tenancy––rules that had been further complicated with the
introduction of the bound colonate––in favor of alternative modes of labor
organization, especially slavery. Where tenancy could still continue on its
own momentum, as on the fundus Tuletianos, whose tenants still
benefitted from the advantages offered by the ongoing maintenance of
their Mancian leases, it did so. But the complexities of the bound colonate
and the restrictions it placed on the freedom of tenants had been furnishing
an incentive to flight already for a century when the Vandals arrived, and
the much looser fiscal and normative apparatus that the Vandals then
introduced had opened the door to a mass exodus of coloni from their
bound tenancies.
The efforts by Justinian to prop up this system were thus unlikely to
have enjoyed much success. A law of Justin II dated to 570 shows
evidence of the ongoing struggle to hold together a system of fiscal
registration that had largely collapsed.110 It contends that African estates
had been deserted of their owners because of a law granting freeborn
108
Nov. Just. App. 6 (Schöll / Kroll p. 799).
109
Nov. Just. App. 9 (Schöll / Kroll p. 803).
110
Nov. Post. Just. Coll. 1 No. 6 (570, Mar. 1) (Zachariä von Lingenthal ed., at
Zepos and Zepos 1: 10–11).
148 Chapter Six
status to any children born of a freeborn woman and a male adscripticius

(registered colonus). The law attempts to correct this problem by
permitting such children to remain free of colonus status but insisting that
they still abide as laborers on the farms on which they had been born. This
new solution would obviously have had limited effect, a fact confirmed by
the issuance of a rescript by Maurice Tiberius in 582 reasserting the old
principle that the children of a free woman and a male colonus were to be
born coloni.111 The bound colonate had thus unraveled into chaos, even if
the east Roman state continued to struggle mightily to hold it together.112
Meanwhile evidence for the use of slaves in this period is nowhere to be
found. The source record for Byzantine North Africa is thin enough that
this should not be taken as an indication of the absence of slaves.
Nevertheless, it does reconfirm the impression made abundantly clear
throughout this study that slavery always played a secondary role behind
tenancy in agricultural production in Roman North Africa.
Conclusion
Beyond this series of sixth–century legal pronouncements we can no
longer say precisely what became of a system that was clearly straining
under its own weight. Whether the bound colonate and its attendant fiscal
structures survived to the eve of the Arab conquest remains unknown.
What is certain is that the penchant toward peasant tenancy rather than
slave labor had a long and enduring history in North Africa. In only two of
the six centuries covered in this study do we have relatively abundant
evidence for the use of agricultural slavery, the second and the later fifth–
early sixth. Even in these periods, however, tenancy is well attested in
solid documentary sources that make it clear that it certainly competed
both demographically and economically with slavery as a reliable
generator of surplus.
Nowhere is the heavy reliance on tenancy more apparent than in the
early fifth century world of Augustine, precisely the context Harper has
singled out as a bellwether of a thriving “slave society” in the late antique
West. Augustine’s high resolution details on the personnel of numerous
early fifth-century farmsteads leaves no room for doubt that tenancy far
111
Nov. Post. Just. Coll. 1 No. 13 (582, Aug. 11) (Zachariä von Lingenthal ed., at
Zepos and Zepos 1: 24).
112
Fentress et al. 2004 note a general decline in archaeological evidence for
settlement in North African in the sixth century.
outpaced slavery as the primary mode of labor organization in this context.

The fact that Augustine was operating in an environment in which the
bound colonate was already fully fledged allows us, however, to catch a
glimpse of the strains this new system of fiscal registration and personal
limitation was putting on social and economic relations. The arrival of the
Vandals, who appear to have relied more heavily on slave labor than their
Roman predecessors, further pointed to cracks in a system so complex and
restrictive that it could only survive in an environment with tight state
control. Thus, when the Roman state regained western North Africa in the
mid fifth century and central North Africa in the mid sixth, it worked to
impose a reversion to the old model of bound tenancy, but the
complications and weaknesses of the late Roman colonate remain evident
in our limited source record. Some version of bound tenancy would of
course continue to be maintained in the Byzantine East down to the
fifteenth century, but North African is more likely gradually to have
settled back into the less restrictive—but also more productive—forms of
tenancy developed already in the first and second centuries CE that had
helped foster the economic boom of the second through fifth centuries.
Ultimately, Africa remained throughout its history a land of agricultural
tenancy with relatively self-assertive peasant laborers. The Romans appear
not only to have accepted this reality but also to have fostered it, for the
liberal land tenure arrangements developed through the lex Manciana
made tenancy more adaptable and productive than slavery in the
Maghreb’s highly varied landscape with its limited water resources. While
some landholders certainly deployed slaves on their estates, they appear
never to have attempted to shift the economy toward productive strategies
that relied predominantly on slave labor. Instead, tenancy is attested over
and over again as the primary mode of organizing farm labor, while there
is essentially no evidence that Roman North Africa was ever a “slave
society” in anything like the sense described by Finley.
Bibliography
Addyman, Peter. 1962. “The Archaeology of the Sbeitla Area: an Interim
Report.” Brathay Exploration Group, Annual Report 1962, 60–77.
Banaji, Jairus. 1997. “Lavoratori Liberi e Residenza Coatta: Il Colonato
Romano in Prospettiva Storica.” In Lo Cascio, Elio, ed. 1997. 253–80.
—. 2001. Agrarian Change in Late Antiquity: Gold, Labour, and
Aristocratic Dominance. New ed. Oxford Classical Monographs.
Oxford: Oxford University Press.
150 Chapter Six
Ben Baaziz, Sadok. 1986. “L'occupation humaine dans la plaine de Rohia

et le Sraa Ouertane dans l'antiquité.” In Histoire et archéologie de
l'Afrique du Nord. Actes du IIIe Colloque international réuni dans le
cadre du 110e Congrès national des Sociétés savantes, Montpellier 1-5
avril 1985, 289–300. Paris: CFHS.
Blazquez, J.M. 1998. “Representaciones de esclavos en mosaicos
africanos.” AfrRom 12: 1029–36.
Campbell, Brian. 2000. The Writings of the Roman Land Surveyors:
Introduction, Text, Translation, and Commentary. Hertford: Society
for the Promotion of Roman Studies.
Carlsen, Jesper. 1991. “Estate Management in Roman North Africa:
Transformation or Continuity? Africa Romana 8: 624-37.
Carrié, Jean-Michel. 1982. “Le ‘Colonat du Bas-Empire’: Un mythe
historiographique.” Opus: International Journal for Social and
Economic History of Antiquity 1: 351–70.
—. 1983. “Un roman des origines: Les généalogies du ‘colonat’ du Bas-
Empire.” Opus: International Journal for Social and Economic
History of Antiquity 2: 205–51.
—. 1997. “‘Colonato del Basso Impero’: La Resistenza Del Mito.” In Lo
Cascio, Elio, ed. 1997, 75–150.
Conant, Jonathan. 2012. Staying Roman: Conquest and Identity in Africa
and the Mediterranean, 439-700. Cambridge: Cambridge University
Press.
Courtois, Christian et al., eds. 1952. Les Tablettes Albertini. Actes privés
de l'époque vandale (fin du Ve siècle). Paris: Arts et métiers
graphiques.
Cracco Ruggini, Lellia. 1987. “‘Coloni’ e ‘Inquilini’: ‘Miseri et egeni
homines’?” AARC 8: 199–216.
Crawford, Dorothy J. 1976. “Imperial Estates.” In Finley, Moses. Studies
in Roman Property, 35-70. Cambridge: Cambridge University Press.
De Ligt, Luuk. 1998–1999. “Studies in Legal and Agrarian History I: the
Inscription from Henchir-Mettich and the Lex Manciana.” Ancient
Society 29: 219–39.
De Vos, Mariette. 2000. Rus Africum. Terra, acqua, olio nell’Africa
settentrionale, Scavo e ricognizione nei dintorni di Dougga (Alto Tell
tunisino). Trento: Università degli Studi di Trento.
Desanges, Jehan. 1988. “Saltus et vicus P(h)osphorianus en Numidie.”
Africa Romana 6: 283–91.
Dietz, S., L. Ladjimi Sebaï, and H. Ben Hassen, eds. 1995. Africa
Proconsularis: Regional Studies in the Segermes Valley of Northern
Tunesia. 2 Vols. Copenhagen: Carlsberg Foundation.
Dossey, Leslie. 2010. Peasant and Empire in Christian North Africa.

Berkeley: University of California.
Dunbabin, Katherine. 2003. “The Waiting Servant in Later Roman Art.”
AJPh 124: 443–68.
Fentress, Elizabeth, et al. 2004. “Accounting for ARS: Fineware and Sites
in Sicily and Africa.” In Alcock, Susan E, and Cherry, John F. Side-by-
side survey: comparative regional studies in the Mediterranean World,
147–62. Oxford: Oxford University Press.
Finley, Moses I. 1973. The Ancient Economy. Berkeley: University of
California.
—. 1980. Ancient Slavery and Modern Ideology. Edited by B.D. Shaw.
Expanded ed. Princeton, NJ [1998]; Princeton University Press.
Flach, Dieter. 1978. “Inschriften Untersuchungen zum römischen Kolonie
in Nordafrika.” Chiron 8: 441–92.
—. 1982. “Die Pachtbedingungen der Kolonen und die Verwaltung der
kaiserlichen Güter in Nordafrika.” ANRW II.10.2: 427–73.
Freu, Christel. 2013. “Les salariés de la terre dans l’antiquité tardive.”
AnTard 21: 283-98.
Fustel de Coulanges, Numa Denis. 1894. Recherches sur quelques
problèmes d’histoire. 2e éd. Paris: Librairie Hachette.
Garnsey, Peter. 1998. “Slavery as Institution and Metaphor in the New
Sermons (with special reference to Dolbeau 2 and 21).” In Madec,
Goulven, ed. 1998. Augustin prédicateur (395–411), Actes du
Colloque International de Chantilly, 471–479. Paris: Collection des
études augustiniennes.
Génelle, Gérard. 2005. La vie économique et sociale dans l'Afrique
romaine tardive d'après les sermons de saint Augustin. Lille: ANRT.
George, Michele G., ed. 2012. Roman Slavery and Roman Material
Culture. Phoenix Supplementary Volumes 52. Toronto: University of
Toronto Press.
Giliberti, Giuseppe. 1999. Servi della terra: ricerche per una storia del
colonato. Turin: Giappichelli.
Grey, Cam. 2011. Constructing Communities in the Late Roman
Countryside. Cambridge: Cambridge University Press.
—. “Contextualizing Colonatus: The Origo of the Late Roman Empire.”
JRS 97 (2007): 155–77.
Gsell, Stephane. 1932. “Esclaves ruraux dans l’Afrique romaine.” In
Mélanges Gustave Glotz, vol. 1, 397–415. Paris: Presses Universitaires
de France.
152 Chapter Six
Harper, Kyle. 2011. Slavery in the Late Roman World, AD 275–425: An

Economic, Social, and Institutional Study. Cambridge: Cambridge
University Press.
—. 2012. “The Transformation of Roman Slavery: An Economic Myth?”
AnTard 20: 165–72.
Hickey, Tod M. 2007. “Aristocratic Landholding and the Economy of
Byzantine Egypt.” In Bagnall, Roger, ed. 2007. Egypt in the Byzantine
World, 450–700, 288–308. Cambridge: Cambridge University Press.
—. 2012. Wine, Wealth and the State in Late Antique Egypt: The House of
Apion at Oxyrhynchus. Ann Arbor, MI: University of Michigan Press.
Hitchner, Robert B. 1988. “The Kasserine Archaeological Survey, 1982–
1986.” AntAfr 24: 7–41.
Hobson, Matthew S. 2015. The North African Boom. Evaluating Economic
Growth in the Roman Province of Africa Proconsularis (146 B.C. –
A.D. 439). JRA Supplements 100. Portsmouth: Rhode Island.
Johne, Klaus-Peter, Köhn, Jens, and Weber, Volker. 1983. Die Kolonen in
Italien und den Westlichen Provinzen des römischen Reiches. Schriften
zur Geschichte und Kultur der Antike 21. Berlin: Akademie Verlag.
Kehoe, Dennis P. 1984a. “Lease Regulations for Imperial Estates in North
Africa. Part I.” Zeitschrift für Papyrologie und Epigraphik 56: 193–
219.
—. 1984b. “Private and Imperial Management of Roman Estates in North
Africa.” Law and History Review 2.2: 241–63.
—. 1985. “Lease Regulations for Imperial Estates in North Africa. Part
II.” Zeitschrift für Papyrologie und Epigraphik 59: 151–72.
—. 1988. The Economics of Agriculture on Roman Imperial Estates in
North Africa. Göttingen: Vandenhoeck & Ruprech.
—. 1997. Investment, Profit, and Tenancy: The Jurists and the Roman
Agrarian Economy. Ann Arbor, MI: University Of Michigan Press.
—. 2008. Law and Rural Economy in the Roman Empire. Ann Arbor, MI:
University of Michigan Press.
Klein, Richard. 1988. Die Sklaverei in der Sicht der Bischöfe Ambrosius
und Augustinus. Forschungen zur antiken Sklaverei 20. Stuttgart:
Steiner.
Kolendo, Jerzy. 1991[1976]. Le colonat en Afrique sous le Haut–Empire,
2nd ed. Paris: Les Belles Lettres.
Lancel, Serge, ed. and trans. 2002. Victor de Vita, Histoire de la
persécution vandale en afrique. Paris: CUF.
Lassère, Jean-M. 1977. Ubique populus: peuplement et mouvements de
population dans l'Afrique romaine de la chute de Carthage à la fin de
la dynastie des Sévères (146 a.C.–235 p.C.). Paris: Éditions du CNRS.
—. 2015. Africa, quasi Roma. 256 av. J.-C–711 ap. J.-C. Paris: Éditions
du CNRS.
Lengrand, Denis. 1996. “Les notables et leur propriétés: La formule ‘in his
praediis’ dans l’Empire romain.” REA 98: 109–31.
Lepelley, Claude. 1967. “Declin ou stabilité de l’agriculture africaine au
Bas–Empire? A propos d’une loi de l’Empereur Honorius.” AntAfr 1:
135–44.
—. 1983. “Liberté, colonat, et esclavage d’après la lettre Divjak 24*
d’Augustin: La jurisdiction épiscopale de liberali causa.” In Les lettres
de Saint Augustin découvertes par Johannes Divjak. Actes du colloque
réuni à Paris les 20 et 21 septembre 1982, 329–42. Paris: Collection
des études augustiniennes.
—. 1989. “Trois documents méconnus retrouvés parmi les spuria de
Sulpice Sévère.” AntAfr 25: 235–62.
—. 1994. “La création de cités nouvelles en Afrique au Bas-Empire: Le
cas de la civitas Faustianensis,” In Yann Le Bohec, ed. L’Afrique, la
Gaule, la Religion à l’époque romaine: Mélanges à la mémoire de
Marcel Le Glay, 288-99. Brussels: Latomus.
Leveau, Philippe. 1984. Caesarea de Maurétanie: Une ville romaine et ses
campagnes. Rome: École française de Rome.
Lo Cascio, Elio. 1997. Terre, Proprietari e Contadini dell’Impero
Romano. Dall’affitto Agrario al Colonato Tardoantico (Incontro
Studio di Capri, 16–18 Ottobre 1995). Rome: NIS.
Mattingly, David J. 1988. “Oil for export? A comparison of Libyan,
Spanish and Tunisian Olive Oil Production in the Roman Empire.”
JRA 1: 33–56.
—. “Olive Oil Production in Roman Tripolitania.” In Buck, David J., and
Mattingly, David J., eds. 1985. Town and Country in Roman
Tripolitania, Papers in honor of O. Hackett, 27–46. Oxford: Oxford
University Press.
Mattingly, David J. et al. 2001. “Leptiminus (Tunisia): A ‘Producer’
City?.” In Mattingly David J., and Salmon, John. 2001. Economies
Beyond Agriculture in the Classical World, 66–89. London: Routledge.
Mattingly, David J. and Hitchner, R. Bruce. 1995. “Roman Africa: An
Archaeological Review.” JRS 85: 165-213.
Mattingly, David J., Sterry, Martin, and Leitch, Victoria. 2013. “Fortified
Farms and Defended Villages of Late Roman and Late Antique
Africa.” AnTard 21: 167–88.
Mazzarino, Santo. 1951. Aspetti sociali del IV secolo. Ricerche di storia
tardo–romana. Rome: «L’Erma» di Bretschneider.
154 Chapter Six
M'Charek, Ahmed. 2003. “Civitas Faustianensis en Byzacène: Un ancien

domaine de Q. Anicius Faustus.” In Bejaoui, Fati, ed. 2003. Histoire
des Hautes Steppes. Antiquité, Moyen âge. Actes du colloque de
Sbeitla, Session 2001, 27–29. Tunis: Institut National du Patrimoine.
Merrills, Andy, and Miles, Richard. 2010. The Vandals. Malden, MA and
Oxford: Wiley–Blackwell.
Mirkovic, Miroslava. 1997. The Later Roman Colonate and Freedom.
Philadelphia: American Philosophical Society.
Modéran, Yves. 2002. “L’établissement territorial des Vandales en
Afrique.” AnTard 10: 87–122.
—. 2014. Les Vandales et l'Empire romain. Arles: Editions Errance.
Peyras, Jean. 1975. “Le Fundus Aufidianus : étude d'un grand domaine
romain de la région de Mateur (Tunisie du Nord).” AntAfr 9: 181–222.
—. 1983. “Paysages agraires et centuriations dans le bassin de l'oued Tine
(Tunisie du Nord).” AntAfr 19: 209–53.
Picard, Gilbert-Charles. 1959. La Civilisation de l’Afrique romaine. Paris:
Picard.
Rosafio, Pasquale. 2002. Studi sul colonato. Bari: Edipuglia.
Rostovtzeff, Mihail I. 1910. Studien zur Geschichte des römischen
Kolonates. Leipzig and Berlin: B. G. Teubner.
—. 1957. Social and Economic History of the Roman Empire, 2nd ed.
Oxford: Clarendon Press.
Sarris, Peter. 2006. Economy and Society in the Age of Justinian.
Cambridge: Cambridge University Press.
Štaerman, Elena M. 1957. Sklaverei in den westlichen Provinzen des
Römischen Reiches im 1.–3. Jahrhundert, trans. J. Kriz. 1957[1987]:
Franz Steiner.
Scholl, Reinhold, and Charlotte Schubert. 2004. “Lex Hadriana de agris
rudibus und Lex Manciana.” APF 50: 79-84.
Schubert, Charlotte. 2008. “Die kaiserliche Agrargesetzgebung in
Nordafrika von Trajan bis Justinian.” ZPE 167: 251–75.
Schulten, Adolf. 1896. Die römischen Grundherrschaften. Eine
agrarhistorische Untersuchung. Weimar: E. Felber.
Shaw, Brent. D. 1982. “Lamasba: An Ancient Irrigation Community.”
AntAfr 18: 61–103.
—. 2011. Sacred Violence: African Christians and Sectarian Hatred in the
Age of Augustine. Cambridge: Cambridge University Press.
—. 2013. Bringing in the Sheaves: Economy and Metaphor in the Roman
World. Toronto: University of Toronto Press.
Sirks, Boudwijn J.A. 2008. “The Colonate in Justinian’s Reign.” JRS 98:
120–43.
Vera, Domenico. 1983. “Strutture agrarie e strutture patrimoniali nella

tarda antichità: l’aristocrazia romana fra agricoltura e commercio.”
Opus 2 (2): 489–533.
—. 1986. “Simmaco e le sue proprietà: struttura e funzionamento di un
patrimonio aristocratico del quarto secolo d.C.” In Paschoud, François,
ed. Colloque genevois sur Symmaque a l’occasion du mille six
centième anniversaire du conflit de l’autel de la Victoire, 231–70.
Paris: Société d’édition Les Belles Lettres”.
—. 1987. “Einfiteusi, colonato e trasformazioni agrarie nell’Africa
Proconsulare nel Tardo Impero.” AfrRom 4: 267–93.
—. 1988. “Terra e lavoro nell’Africa romana.” Studi Storici 29 (4): 967–
92.
—. 1992. “Conductores domus nostrae, conductores privatorum:
Concentrazione fondiaria e redistribuzione della ricchezza nell’Africa
tardoantica.” In Christol, Michel et al., eds. 1992. Institutions, société
et vie politique dans l’empire romain au IVe siècle ap. J.-C., 465–90.
Rome: École française de Rome.
—. 1992–1993. “Schiavitù rurale e colonato nell’Italia Imperiale.” Scienze
dell’Antichità: Storia Archeologia Antropologia 67: 291–339.
—. 2012. “Questioni di storia agraria tardoromana: Schiavi, Coloni,
Villae.” AnTard 20: 115–122.
Weber, Max. 1891. Roman Agrarian History in Its Relation to Roman
Public & Civil Law. Translated by Frank Richard I. Claremont, CA:
Regina Books [repr. 2008].
Wessel, Hendrik. 2003. Das Recht der Tablettes Albertini. Freiburger
Rechtsgeschichtliche Abhandlungen 40. Berlin: Duncker & Humblot.
Wickham, Chris. 2005. Framing the Early Middle Ages, 400–800. Oxford:
Oxford University Press.
Witschel, Christian. 1999. Krise––Rezession––Stagnation? Der Westen
des römischen Reiches im 3. Jahrhundert n. Chr. Frankfurt am Main:
Buchverlag Marthe Clauss.
Whittaker, Charles R. 1978. “Land and Labour in North Africa.” Klio 60:
331–62.
—. 1980. “Rural Labour in Three Roman Provinces.” In Garnsey, Peter.
1980. Non–Slave Labour in the Greco–Roman World, edited by 73–99.
Cambridge: Cambridge University Press.

Lenski Peasant - and - Slave - in - Late - Antique - North

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lenski Peasant - and - Slave - in - Late - Antique - North

Uploaded by

Copyright:

Available Formats

Late Antiquity in

Rita Lizzi Testa

Edited by Rita Lizzi Testa

This book first published 2017

Cambridge Scholars Publishing

Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK

British Library Cataloguing in Publication Data

Copyright © 2017 by Rita Lizzi Testa and contributors

ISBN (10): 1-4438-4308-3

International Committee of Historical Sciences –

Introduction ............................................................................................... vii

Chapter One ................................................................................................. 2

Chapter Two .............................................................................................. 15

Methodology: Sources and Periodization

Chapter Three ............................................................................................ 38

Chapter Four .............................................................................................. 56

Chapter Five .............................................................................................. 80

Chapter Six .............................................................................................. 113

Chapter Seven.......................................................................................... 156

Chapter Eight ........................................................................................... 174

Contributors ............................................................................................. 228

PEASANT AND SLAVE IN LATE ANTIQUE

society” throughout the long fourth century.3 Engaging in particular with

Natural parameters––land and water, large estates

precious supplies of water using dams, reservoirs, and conduits. This

In Africa, then, private estates were often as extensive as neighboring

These competing narratives are in many ways emblematic of a broader

seasonality played a central role in the agricultural labor cycle.20 Finally,

First and second centuries

(Medjerda) River Valley in the province of Africa Proconsularis.22 All are

land (subseciva)––were part of the original lex Manciana, or whether the

apparently were cross–applied throughout North Africa, where they

tenancy arrangement) continue to record the claims of tenants to the

rural estates, but these are far outnumbered by the 66 inscriptions

much in contrast with traditional models of third–century collapse.54

Herodian describes the rebel band as consisting of “those cultivating the

relationship might well have strengthened the position of these

(origo) in perpetuity. Naturally, the aspirations of the state to lock tenants

• First, it responded to problems that arose from attempts to

2. The status of colonus became heritable across generations (CJ

Thus, regardless of where one stands on the question of whether the

Expositio useless as a source for agricultural slavery in the Maghreb.66 In a

interchangeable on imperial estates. Thus a law of 383 addressed to an

Early fifth century

named Donatus, the former proconsul Africae, to enforce the conversion to

for cultivation in addition to the variety of domestic tasks assigned them.85

composition of the Gospels.88 But in descriptions focusing on the

by the Vandals as a standard form of punishment;92 slaves were also

There are, however, clearer indications of the ongoing importance of

And in the province of Sitifensis and Caesariensis, estates under

Valentinian was thus struggling to re-impose upon imperial holdings as

words, the western emperor wished to revive the system of elite

status to any children born of a freeborn woman and a male adscripticius

outpaced slavery as the primary mode of labor organization in this context.

Ben Baaziz, Sadok. 1986. “L'occupation humaine dans la plaine de Rohia

Dossey, Leslie. 2010. Peasant and Empire in Christian North Africa.

Harper, Kyle. 2011. Slavery in the Late Roman World, AD 275–425: An

M'Charek, Ahmed. 2003. “Civitas Faustianensis en Byzacène: Un ancien

Vera, Domenico. 1983. “Strutture agrarie e strutture patrimoniali nella

You might also like